International Business Machines Corporation

Patent Trials and Appeals Board

Dec 2, 2021

2020005211 (P.T.A.B. Dec. 2, 2021)

UNITED STATES PATENT AND TRADEMARK OFFICE
UNITED STATES DEPARTMENT OF COMMERCE
United States Patent and Trademark Office
Address: COMMISSIONER FOR PATENTS
P.O. Box 1450
Alexandria, Virginia 22313-1450
www.uspto.gov
APPLICATION NO. FILING DATE FIRST NAMED INVENTOR ATTORNEY DOCKET NO. CONFIRMATION NO.
15/339,973 11/01/2016 Corville O. Allen SVL920160169US1 5108
45725 7590 12/02/2021
Walder Intellectual Property Law PC
445 Crestover Circle
Richardson, TX 75080
EXAMINER
SASS, KIMBERLY A.
ART UNIT PAPER NUMBER
3626
MAIL DATE DELIVERY MODE
12/02/2021 PAPER
Please find below and/or attached an Office communication concerning this application or proceeding.
The time period for reply, if any, is set in the attached communication.
PTOL-90A (Rev. 04/07)

UNITED STATES PATENT AND TRADEMARK OFFICE
BEFORE THE PATENT TRIAL AND APPEAL BOARD
Ex parte CORVILLE O. ALLEN, TIMOTHY A. BISHOP,
ALBERT A. CHUNG, and ELIZABETH A. SCHREIBER
Appeal 2020-005211
Application 15/339,973
Technology Center 3600
Before DONALD E. ADAMS, TAWEN CHANG, and
RACHEL H. TOWNSEND, Administrative Patent Judges.

CHANG, Administrative Patent Judge.
DECISION ON APPEAL
Pursuant to 35 U.S.C. § 134(a), Appellant1 appeals from the
Examiner’s decision to reject claims 1, 10, 11, 20–36. We have jurisdiction
under 35 U.S.C. § 6(b).
We AFFIRM.

1 We use the word “Appellant” to refer to “applicant” as defined in 37
C.F.R. § 1.42. Appellant identifies the real party in interest as International
Business Machines Corporation. Appeal Br. 2.
Appeal 2020-005211
Application 15/339,973

2
STATEMENT OF THE CASE
The Specification states that “[r]ecent trends . . . are to collect patient
information from a variety of sources into Health Information Exchanges
(HIEs).” Spec. ¶ 19. The Specification states that
[t]his collection of patient medical information, or patient
medical data, from disparate computing systems leads to
many opportunities to improve the care being provided to
patients, one opportunity being the ability to reconcile
which medications a patient is truly taking, and should be
taking, so as to improve the quality of care provided to the
patient and inform health providers of the medication
status of the patient so that informed decisions may be
made.
Id. ¶ 23. According to the Specification, the invention “relates generally to
an improved data processing apparatus and method and more specifically to
mechanisms for providing cognitive medication reconciliation for use with
cognitive systems,” e.g., “decision support systems” such as “a diagnosis
system employed in the healthcare industry.” Id. ¶¶ 1–2.

CLAIMED SUBJECT MATTER
The claims are directed to a method for processing a plurality of
patient medical data received from a plurality of different source computing
systems, and computer program product and apparatus for performing the
same. Claim 1 is illustrative:
1. A method, in a data processing system comprising at
least one processor and at least one memory, the at least one
memory comprising instructions executed by the at least one
processor to cause the at least one processor to be specifically
configured to execute the operations of the method in the data
processing system, comprising:
Appeal 2020-005211
Application 15/339,973

3
configuring the at least one processor of the data
processing system, based on the execution of instructions in the
at least one memory, to implement a cognitive computing
system and a patient information aggregation system
comprising a medication reconciliation engine that processes a
plurality of patient medical data, received from a plurality of
different source computing systems;
receiving, by the patient information aggregation system
executing on the data processing system, the plurality of patient
medical data for a patient from the plurality of different source
computing systems;
analyzing, by the cognitive computing system executing
on the data processing system, the plurality of patient medical
data to identify the medication related content within the
plurality of patient medical data at least by performing
computer executed natural language processing on the plurality
of patient medical data to extract portions of natural language
text corresponding to medications specified in the plurality of
patient medical data;
generating, by the medication reconciliation engine
executing on the data processing system, an aggregate
medication listing data structure for the patient from the
medication related content identified in the plurality of patient
medical data;
correlating, by the medication reconciliation engine
executing on the data processing system, medication related
data types, among the medication related content within the
plurality of patient medical data, which are related to a same
medication or class of medication;
determining, by medication list update logic of the
medication reconciliation engine executing on the data
processing system, whether a modification to the aggregate
medication listing data structure is to be performed based on
results of the correlation; and
outputting, by the patient information aggregation system
executing on the data processing system, a notification to a
computing device associated with an authorized user indicating
a recommended modification to the aggregate medication
listing data structure, in response to determining that a
Appeal 2020-005211
Application 15/339,973

4
modification is to be performed wherein determining whether a
modification to the aggregate medication listing data structure
is to be performed based on results of the correlation further
comprises:
classifying medications specified in the aggregate
medication listing data structure into corresponding classes of
medication based on the related data types; and
determining, for each medication in a same class of
medication in the medication listing data structure, whether the
medication is a valid duplicate or an invalid duplicate
medication relative to one or more other medications in the
same class of medication based on an application of evaluation
rules implemented in logic of the duplicate medication analysis
engine to characteristics of the medication and characteristics of
the one or more other medications in the same class, wherein a
valid duplicate is a duplicate that is determined should remain
in the aggregate medication listing data structure, and an invalid
duplicate is a duplicate that is determined should be removed
from the aggregate medication listing data structure.
Appeal Br. 60–61 (Claims App.).

REJECTION(S)
A. Claims 1, 10, 11, and 20–36 are rejected under 35 U.S.C. § 101 as
being directed to a judicial exception to patent-eligible subject matter,
without significantly more. Final Act. 6.
B. Claims 1, 10, 11, 20, 21, 25, 26, 29, 33, and 34 are rejected under 35
U.S.C. § 103 as obvious2 over Miller3 and Boone.4 Final Act. 10.

2 The Examiner described each of rejections B, C, and D as an anticipation
rejection under 35 U.S.C. § 103 in paragraphs 24, 32, and 38 of the Final
Action. Final Act. 10, 23, 34. We understand these to be typographical
errors and treat the rejections as obviousness rejections.
3 Miller et al., US 2008/0015894 A1, published Jan. 17, 2008.
4 Boone et al., US 2004/0243545 A1, published Dec. 2, 2004.
Appeal 2020-005211
Application 15/339,973

5
C. Claims 22, 23, 27, 28, 30, 31, 35, and 365 are rejected under 35 U.S.C.
§ 103 as obvious over Miller, Boone, and Biernacki.6 Final Act. 23.
D. Claims 24 and 32 are rejected under 35 U.S.C. § 103 as obvious over
Miller, Boone, Biernacki, and Arshad.7 Final Act. 34.
E. Claims 22–24, 27, 28, 30–32, 35, and 36 are rejected under 35 U.S.C.
§ 112(a) or § 112, first paragraph, as failing to comply with the
written description requirement. Final Act. 2–3.

OPINION
A. Rejection under 35 U.S.C. § 101 (claims 1, 10, 11, 20–36)
1. Issue
The Examiner concludes that the claims recite limitations that, “under
their broadest reasonable interpretation, cover certain methods of organizing
human activity by managing interactions between a patient and user to
cognitively reconcile medication lists.” Final Act. 6. Accordingly, the
Examiner asserts that these limitations fall within one of the groupings of
abstract ideas enumerated in the “2019 Revised Patent Subject Matter
Eligibility Guidance” (“Revised Guidance”) issued by the Director of the
USPTO on January 7, 2019. Id. at 6–7.

5 The Examiner did not include claim 36 in the list of claims rejected as
obvious over Miller, Boone, and Biernacki. Final Act. 23. However, in his
analysis he notes that claim 36 is “significantly similar” to claim 28 and is
rejected upon the same art. Accordingly, we understand the obviousness
rejection over Miller, Boone, and Biernacki to include claim 36.
6 Biernacki et al., US 2014/0316797 A1, published Oct. 23, 2014.
7 Arshad et al., US 2017/0011195 A1, published Jan. 12, 2017.
Appeal 2020-005211
Application 15/339,973

6
The Examiner states that the claims do not integrate the abstract idea
into a practical application. Final Act. 7. In particular, the Examiner notes
that the limitations in the claim, other than those reciting abstract ideas, are
“recited at a high level of generality” and/or “amount[s] to no more than
mere instructions to apply the [abstract ideas] using generic computer
components,” and thus “do not impose any meaningful limits on practicing
the abstract idea[s].” Id.
Finally, the Examiner finds that “[t]he claims do not include
additional elements that are sufficient to amount to significantly more than
the judicial exception,” because the additional elements, including when
viewed as an ordered combination, are merely “generic components that are
configured to perform well-understood, routine, and conventional activities
previously known to the industry” and/or “instructions to apply [a judicial]
exception using generic component[s].” Final Act. 7–9.
Appellant contends that the claimed invention does not recite a
judicial exception such as an abstract method of organizing human activity
or mental process. Appeal Br. 18, 21–22, 24–27. Appellant contends that,
instead, the claimed invention is “clearly a practical application,” because
the claim is directed to “an improved computer tool for improving the
functionality of computer cognitive systems applied to medical cognitive
operations specifically by identifying and reconciling medication listings for
patients upon which the computer cognitive systems operate.” Appeal Br.
22, 24; see also id. at 27–28, 30. Finally, Appellant contends that the
computer functionalities recited in the independent claims are not generic
computer functions, that the claims impose “meaningful limits” on any
Appeal 2020-005211
Application 15/339,973

7
alleged judicial exception, and that the Examiner’s rejection amounts to a
position that “software is not patentable per se.” Appeal Br. 28–29.
The issue with respect to this rejection is whether a preponderance of
evidence of record supports the Examiner's determination that the claims are
directed to a judicial exception to patent-eligible subject matter, without
significantly more.
2. Analysis
We analyze this case under the framework set forth by the Supreme
Court in Mayo Collaborative Services v. Prometheus Laboratories, Inc., 566
U.S. 66 (2012), and applied by our reviewing court in Ariosa Diagnostics,
Inc. v. Sequenom, Inc., 788 F.3d 1371 (Fed. Cir. 2015). As the Ariosa court
explained:
In Mayo . . . , the Supreme Court set forth a framework for
distinguishing patents that claim laws of nature, natural
phenomena, and abstract ideas from those that claim
patent-eligible applications of those concepts. First, we
determine whether the claims at issue are directed to a
patent-ineligible concept. . . . If the answer is yes, then
we next consider the elements of each claim both
individually and “as an ordered combination” to determine
whether additional elements “transform the nature of the
claim” into a patent-eligible application. . . . The Supreme
Court has described the second step of this analysis as a
search for an “inventive concept”—i.e., an element or
combination of elements that is “sufficient to ensure that
the patent in practice amounts to significantly more than a
patent upon the [ineligible concept] itself.”
Id. at 1375.
Whether Claims Are Directed to Patent-Ineligible Concept
We begin with the first step of the Mayo test, namely whether a claim
is “directed to” a patent-ineligible concept. On January 7, 2019, the Director
Appeal 2020-005211
Application 15/339,973

8
of the USPTO issued the “2019 Revised Patent Subject Matter Eligibility
Guidance” (“Revised Guidance”), which provides further details regarding
how the Patent Office analyzes patent-eligibility questions under 35 U.S.C.
§ 101. 84 Fed. Reg. 50–57 (Jan. 7, 2019). Under the Revised Guidance, the
first step of the Mayo test (i.e., Step 2A of the Revised Guidance)8 is “a two-
pronged inquiry.” Id. at 54. In prong one, we evaluate whether the claim
recites a judicial exception, such as laws of nature, natural phenomena, or
abstract ideas. Id. The Revised Guidance explains that
the abstract idea exception includes the following
groupings of subject matter, when recited as such in a
claim limitation(s) (that is, when recited on their own or
per se):
(a) Mathematical concepts—mathematical
relationships, mathematical formulas or equations,
mathematical calculations;
(b) Certain methods of organizing human activity—
fundamental economic principles or practices (including
hedging, insurance, mitigating risk); commercial or legal
interactions (including agreements in the form of
contracts; legal obligations; advertising, marketing or
sales activities or behaviors; business relations); managing
personal behavior or relationships or interactions between
people (including social activities, teaching, and following
rules or instructions); and

8 As to Step 1 described in the Revised Guidance, there is no dispute that the
subject matters of the claims on appeal “fall[] within the four statutory
categories of patentable subject matter identified by 35 U.S.C. [§] 101:
Process, machine, manufacture, or composition of matter.” 84 Fed. Reg.
53–54.
Appeal 2020-005211
Application 15/339,973

9
(c) Mental processes—concepts performed in the
human mind (including an observation, evaluation,
judgment, opinion).
Id. at 52 (footnotes omitted). The Revised Guidance states that, except in
rare circumstances, “[c]laims that do not recite matter that falls within these
enumerated groups of abstract ideas should not be treated as reciting abstract
ideas” and thus would be patent-eligible. Id. at 53. If the claim recites a
judicial exception, the claim is further analyzed under prong two, which
requires “evaluat[ion of] whether the claim recites additional elements that
integrate the exception into a practical application of that exception.” Id.
The Revised Guidance explains that, “[i]f the recited exception is integrated
into a practical application of the exception, then the claim is eligible at
Prong Two of . . . Step 2A [of the Revised Guidance].” Id.
i. Prong One of Step 2A of Revised Guidance
Appellant does not separately argue independent claims 1, 11, and 20.
We therefore focus our analysis on claim 1 as representative of the
independent claims. We further analyze the patent eligibility of certain
separately argued dependent claims in the portion of the opinion addressing
Appellant’s arguments.
We agree with the Examiner that claim 1 recites patent-ineligible
subject matter. In particular, we find that the claim recites a data processing
system receiving, analyzing, and outputting patient medical data, which
merely automates a well-established method of organizing human activity.
In re Salwan, 681 Fed. App’x 938, 941 (Fed. Cir. 2017) (non-precedential)
(finding claim reciting “using billing software to calculate a patient’s bill
based on [electronic medical records] and insurance information” to be mere
Appeal 2020-005211
Application 15/339,973

10
“automation of a ‘method of organizing human activity’ with respect to
medical information” and, thus, directed to the abstract idea).
Likewise, mental processes are abstract. 84 Fed. Reg. 52. Claim 1
recites (1) “receiving . . . [a] plurality of patient medical data for a patient”;
(2) “analyzing . . . [a] plurality of patient medical data to identify the
medication related content . . . by . . . extract[ing] portions of natural
language text corresponding to medications specified in the . . .patient
medical data”; (3) “generating . . . an aggregate medication listing data
structure for the patient from the medication related content identified in the
plurality of patient medical data”; (4) “correlating . . . medication related
data types, among the medication related content within the plurality of
patient medical data, which are related to a same medication or class of
medication”; (5) “determining . . . whether a modification to an aggregate
medication listing data structure is to be performed based on the results of
the correlation”; (6) “classifying medications specified in the aggregate
medication listing data structure into corresponding classes of medication
based on the related data types”; and (7) “determining . . . whether the
medication is a valid duplicate or an invalid duplicate medication relative to
one or more other medications in the same class of medication . . . .” Appeal
Br. 60–61 (Claims App.). These steps are mental processes — i.e., steps
involving observation, evaluation, and/or judgment that could be “performed
in the human mind, or by a human using pen and paper” — and accordingly
fall within the realm of abstract ideas. CyberSource Corp. v. Retail
Decisions, Inc., 654 F.3d 1366, 1372 (Fed. Cir. 2011) (holding that a claim
is directed to “unpatentable mental processes” where all of the claim’s
Appeal 2020-005211
Application 15/339,973

11
method steps “can be performed in the human mind, or by a human using a
pen and paper”).
ii. Prong Two of Step 2A of Revised Guidance
With respect to the second prong of Step 2A of the Revised Guidance,
we find that claim 1 does not recite sufficient additional elements that
integrate the recited judicial exception(s) into a practical application of the
exception(s). 84 Fed. Reg. at 54. More specifically, the additional elements
of claim 1 that are not either a method of organizing human activity or
mental processes include (1) “a data processing system” and its various
components such as processor, memory, cognitive computing system,
patient information aggregation system, medication reconciliation engine,
and medication list update logic; (2) “instructions [in the memory] executed
by the . . . processor to cause the . . . processor to be specifically configured
to execute the operations of the method in the data processing system”;
(3) the step of “configuring the . . . processor of the data processing system”
to implement the various components of the data processing system; (4) “a
plurality of different source computing systems” from which patient medical
data are received; and (5) “computing devices associated with an authorized
user.” Appeal Br. 60–61 (Claims App.).
These additional elements, individually and in combination, do not
suffice to integrate the recited judicial exceptions into a practical application.
Instead, they merely recite “us[ing] a computer as a tool to perform an
abstract idea,” which does not integrate a judicial exception into a practical
application. 84 Fed. Reg. at 55; See also In re Salwan, 681 Fed. App’x at
941 (mere automation of an abstract idea does not render claim patent
eligible); Intellectual Ventures I LLC v. Symantec Corp., 838 F.3d 1307,
Appeal 2020-005211
Application 15/339,973

12
1314–15 (Fed. Cir. 2016) (explaining, in the context of claims directed to
“filtering e-mails that have unwanted content,” that “it was long-prevalent
practice for people receiving paper mail to look at an envelope and discard
certain letters, without opening them, from sources from which they did not
wish to receive mail based on characteristics of the mail,” and noting that
“appl[ying such] a well-known idea using generic computers ‘to the
particular technological environment of the Internet’” does not render a
claim patent eligible).
Whether Claim 1 Amounts to “Significantly More”
Having determined that claim 1 is directed to a patent-ineligible
abstract idea, we next consider whether claim 1 recites “an element or
combination of elements that is ‘sufficient to ensure that the patent in
practice amounts to significantly more than a patent upon the [ineligible
concept] itself.’” Ariosa, 788 F.3d at 1375 (citation omitted). We agree
with the Examiner that it does not.
As discussed above, the additional elements in claim 1 that are not
either part of a method of organizing human activity or mental processes
include (1) a data processing system and its various components; (2)
“instructions” in the memory to cause the computer processor to execute the
method; (3) the step of configuring the processor to implement the various
components of the data processing system; (4) “a plurality of different
source computing systems” from which patient medical data are received;
and (5) “computing devices associated with an authorized user.” Appeal Br.
60–61 (Claims App.).
As also discussed above, however, these limitations merely recite a
generic computer system performing the generic computer functions of
Appeal 2020-005211
Application 15/339,973

13
receiving, analyzing, and outputting data. As our reviewing court has
explained,
[c]laims that “amount to nothing significantly more than
an instruction to apply [an] abstract idea ... using some
unspecified, generic computer” and in which “each step
does no more than require a generic computer to perform
generic computer functions” do not make an abstract idea
patent-eligible, [Alice Corp. Pty. Ltd. v. CLS Bank Int’l,
573 U.S. 208, 225–226 (Fed. Cir. 2014)] (citations and
internal quotation marks omitted), because “claiming the
improved speed or efficiency inherent with applying the
abstract idea on a computer” does not “provide a sufficient
inventive concept.” Intellectual Ventures I LLC v. Capital
One Bank (USA) (“Intellectual Ventures v. Capital One
Bank”), 792 F.3d 1363, 1367 (Fed. Cir. 2015).
Intellectual Ventures I LLC v. Symantec Corp., 838 F.3d 1307, 1315 (Fed.
Cir. 2016).
Appellant’s Arguments
Appellant contends that “the claimed invention does not involve any
method of organizing human activity, let alone the certain methods of
organizing human activity recognized by the 2019 PEG and the 2019
Updated Guidance for the 2019 PEG.”9 Appeal Br. 21–22; see also id. at
25–26, Reply Br. 6. In particular, Appellant contends that “the claims
cannot be properly construed as being directed to or even including any
management of interactions between a patient and a user,” because
“nowhere in the claims is there any interaction between a patient and a user
recited.” Appeal Br. 18; see also id. at 26.

9 We understand Appellant’s citations to the “2019 PEG” and the “2019
Updated Guidance for the 2019 PEG” to refer to, respectively, the Revised
Guidance and the “October 2019 Update: Subject Matter Eligibility.”
Appeal 2020-005211
Application 15/339,973

14
We are not persuaded. A claim need not specifically recite an
interaction between a patient and a user in order to recite an abstract method
of organizing human activity, because, as our reviewing court has explained,
automation of such interactions remains abstract. See, e.g., In re Salwan,
681 Fed. App’x at 941; Intellectual Ventures I LLC, 838 F.3d at 1314–15.
In this case, it is a long-prevalent practice for medical providers (e.g.,
authorized users in the claimed invention) to query patients regarding
medications they have been prescribed and/or are taking, and for patients to
supply such information, in order for the medical providers to optimize the
medications taken by the patient, including by removing duplications where
necessary. See, e.g., Biernacki ¶ 5 (explaining that “[c]oincident with the
management of medical conditions and administration of medications is the
medication reconciliation and review process,” which “will typically begin
[with] compiling a patient’s current active medication list . . . by either
receiving a list as compiled by a patient or caregiver, and/or via clinician
review”). The fact that the claims automate this process and thus remove
direct interaction between the patient and user does not thereby render the
claims patent eligible.
Appellant similarly contends that the claims recite “an operation that
is not performed in a human mind”—i.e., “identif[ying] medication related
content within the patient medical data at least by performing computer
executed natural language processing . . . to extract portions of natural
language text corresponding to medications,” which Appellant contends to
be “specific to computer technology as natural language processing is a term
of art specific to computer technology.” Id. at 24; see also id. at 26–27.
Appeal 2020-005211
Application 15/339,973

15
We are not persuaded. The step of identifying medication related
content within patient medical data by processing “natural language” to
extract “portions of natural language text corresponding to medications” can
be performed in the human mind, e.g., in the mind of a medical provider
reviewing patient medical data. Neither does the fact the claim recites
“computer executed” natural language processing remove this step from the
realm of abstract mental processes. See, e.g., Intellectual Ventures I LLC,
838 F.3d at 1318 (finding as abstract claims directed to “[a] post office for
receiving and redistributing email messages on a computer network” in part
because, “with the exception of generic computer-implemented steps, there
is nothing in the claims themselves that foreclose them from being
performed by a human, mentally or with pen and paper”).
Appellant contends that “the claimed invention is . . . directed to an
improved computer tool for improving the functionality of computer
cognitive systems applied to medical cognitive operations specifically by
identifying and reconciling medication listings for patients upon which the
computer cognitive systems operate.” Appeal Br. 22; see also id. at 27.
Appellant contends that the invention is “specifically concerned with issues
arising as a result of the advent of health information exchanges (HIE) and
other similar patient information aggregation computer systems” and “solves
the problems with existing medical computing systems which are myopic in
that they only know the information that they themselves maintain and do
not have a complete picture of the patient’s medical condition especially
with regard to medications.” Id. at 20, 23; see also id. at 28, Reply Br. 7–8.
Appellant contends that, “[b]y providing an improved computing system that
is configured to process information from a variety of different sources
Appeal 2020-005211
Application 15/339,973

16
specifically with regard to reconciling medications prescribed to the patient,
the treatment of the patient is improved.” Appeal Br. 23; see also id. at 28.
Accordingly, Appellant contends that the claimed invention “not only
improves the functionality of the computing system by giving it new
functionality that did not exist prior to the presently claimed invention, but
also improves other technology with regard to cognitive medical treatment
recommendation systems and the practice of medicine.” Id. at 28. For these
reasons, Appellant contends that the claimed invention is “clearly a practical
application.” Id. at 24; see also id. at 27–28, 30.
We are not persuaded. The “improvement” Appellant claims — an
allegedly improved way for “identifying and reconciling medication listings
for patients” — is part of the abstract idea (i.e., part of the method of
organizing human activity and/or mental process) — and does not improve
the functioning of the “computer cognitive system” itself. That is, any
potential improvement in speed or efficiency, such as the acquisition of a
more complete or accurate listing of a patient’s medications, is “inherent
with applying the abstract idea on a computer,” which is “insufficient to
render the claims patent eligible as an improvement to computer
functionality.” See, e.g., Customedia Tech., LLC v. Dish Network Corp.,
951 F.3d 1359, 1364 (Fed. Cir. 2020). The claim thus does not recite an
“additional element” – i.e., “claim features, limitations, and/or steps that are
recited in the claim beyond the identified judicial exception” – that integrates
the judicial exception into a practical application. Revised Guidance, 84
Fed. Reg. at 55, n. 24 (emphasis added).
For similar reasons, we are also not persuaded to the extent Appellant
contends that the improvement to the computing system is in the
Appeal 2020-005211
Application 15/339,973

17
configuration of the computing system “to process information from a
variety of different sources specifically with regard to reconciling
medications prescribed to the patient,” thus giving the computing the “new”
functionality of identifying and reconciling medications. Appeal Br. 23, 28.
Processing information from a variety of different sources is a generic
computer function, see, e.g., buySAFE, Inc. v. Google, Inc., 765 F.3d 1350,
1355 (Fed. Cir. 2014) (explaining a limitation “[t]hat a computer receives
and sends . . . information over a network—with no further specification—is
not even arguably inventive”), and the fact that the information here is
specifically related to reconciling medication does not change our
patentability analysis, see, e.g., Intellectual Ventures I LLC v. Capital One
Bank (USA), N.A., 792 F.3d 1363, 1366 (Fed. Cir. 2015) (explaining that
“[a]n abstract idea does not become nonabstract by limiting the invention to
a particular field of use or technological environment, such as the Internet
[or] a computer”). Likewise, Appellant does not persuade us that the
invention “improves other technology with regard to cognitive medical
treatment recommendation systems and the practice of medicine” because
the invention improves the treatment of the patient. Appeal Br. 23, 28.
Appellant has not persuasively shown that any alleged improvement is to a
technology rather than an abstract idea.
Neither do we agree, as Appellant appears to contend, that the claimed
invention provides a solution that arises only in the context of the Internet or
a computer. DDR Holdings, LLC v. Hotels.com, L.P., 773 F.3d 1245, 1258
(Fed. Cir. 2014). Contrary to Appellant’s apparent assertion that the claimed
invention solves a problem “specifically arising as a result of the advent of
. . . patient information aggregation computer systems,” namely “myopic”
Appeal 2020-005211
Application 15/339,973

18
medical computing systems that “only know the information that they
themselves maintain and do not have a complete picture of the patient’s
medical condition especially with regard to medications,” Appeal Br. 20, 23,
28, the problem of a medical provider lacking complete medical information
with respect to a patient is a long standing one that predates, e.g., health
information exchange (HIE). See, e.g., Miller ¶ 4 (explaining that “a patient
may have different medical providers for different medical conditions,
resulting in different medication prescriptions” and that, “[a]s a result, in
many instances the patient, the patient’s medical provider and/or the
patient’s pharmacist is not fully apprised of the patient’s medication therapy
regimen”).
Appellant contends that the computer functionalities recited in the
independent claims are not generic computer functions. Appeal Br. 29.
Appellant contends that the Examiner’s rejection amounts to a position that
“software is not patentable per se.” Id. at 28; see also Reply Br. 8.
We are not persuaded. In particular, the computer functionalities
recited in claim 1 include processing data received from a plurality of
different source computers, analyzing the data to extract information (i.e.,
identify medication related content by performing computer executed natural
language processing), performing additional analysis on the extracted
information, and outputting the results of the analysis. Appeal Br. 60–61
(Claims App.). Contrary to Appellant’s contention, these functions are all
generic computer functions. See, e.g., buySAFE, 765 F.3d at 1355
(receiving and sending information over a network not inventive); In re Bd.
of Trustees of the Leland Stanford Junior Univ., 991 F.3d 1245, 1251 (Fed.
Cir. 2021) (holding that “the recited steps of receiving, extracting, and
Appeal 2020-005211
Application 15/339,973

19
storing data amount to well-known, routine, and conventional steps taken
when executing a mathematical algorithm on a regular computer”).
We agree with Appellant that “software can make non-abstract
improvements to computer technology just as hardware improvements
can.” Enfish, LLC v. Microsoft Corp., 822 F.3d 1327, 1335 (Fed. Cir. 2016).
However, as our reviewing court has explained in the context of whether
claims reciting computer software are patent eligible,
to be directed to a patent-eligible improvement to
computer functionality, the claims must be directed to an
improvement to the functionality of the computer or
network platform itself. See, e.g., id. 1336–39; DDR
Holdings, LLC v. Hotels.com, L.P., 773 F.3d 1245, 1257–
59 (Fed. Cir. 2014). Thus, this inquiry “often turns on
whether the claims focus on ‘the specific asserted
improvement in computer capabilities ... or, instead, on a
process that qualifies as an “abstract idea” for which
computers are invoked merely as a tool.’ ” Finjan, 879
F.3d at 1303 (quoting Enfish, 822 F.3d at 1335–36).
Customedia Tech., 951 F.3d at 1365.
In this case, as discussed above, the focus of the claims is on a method
that qualifies as an abstract idea for which computers are used only as a tool.
For this reason, Appellant’s citations to Enfish, McRO, and Finjan are
inapposite. Reply Br. 8–14 (discussing Enfish, 822 F.3d 1327, McRO, Inc.
v. Bandai Namco Games Am. Inc., 837 F.3d 1299 (Fed. Cir. 2016), and
Finjan Inc. v. Blue Coat Sys., Inc., 879 F.3d 1299 (Fed. Cir. 2018)). The
claims in Enfish are directed to a specific type of data structure (i.e., self-
referential table) “designed to improve the way a computer stores and
retrieves data in memory.” Enfish, 822 F.3d at 1339. Similarly, the claims
in Finjan recited “the use of a ‘behavior-based’ virus scan that was able to
identify and compile unique information about potentially hostile operations,
Appeal 2020-005211
Application 15/339,973

20
while the traditional scan method was limited to recognizing the presence of
previously identified viruses.” Universal Secure Registry LLC v. Apple Inc.,
10 F.4th 1342, 1352 (Fed. Cir. 2021) (citing Finjan, 879 F.3d at 1304).
Thus, unlike the claims on appeal, these claims are directed to improvements
on how the computer itself functions, not merely to a method where
conventional means are used to manipulate data received by and/or stored in
the computer. As for McRO, the claimed improvement in that case was to
“how the physical display operated (to produce better quality images), unlike
(what is present here) a claimed improvement” in the abstract idea of
reconciling medications. SAP America, Inc. v. Investpic, LLC, 898 F.3d
1161, 1167 (Fed. Cir. 2018) (emphasis added).
Finally, Appellant contends that “[t]he present claims clearly impose a
meaningful limit on the alleged judicial exception.” Appeal Br. 29. We are
not persuaded for the reasons discussed above. Accordingly, we affirm the
Examiner’s rejection of claim 1 as being directed to an abstract idea, without
significantly more.
i. Dependent claims
Appellant contends with respect to all of the dependent claims that
they “recite additional features that also demonstrate the patent eligibility of
the claimed invention” and that the Examiner has not established a prima
facie case of patent ineligibility as to these claims because, with respect to
the dependent claims, the Examiner states only that “they do not remedy the
deficiencies of their parent claims,” which “does not demonstrate any actual
consideration of any of the features actually recited in the dependent
claims.” Appeal Br. 31.
Appeal 2020-005211
Application 15/339,973

21
We are not persuaded. Our reviewing court has explained that
[t]he Patent and Trademark Office (“PTO”) satisfies its
initial burden of production by “adequately explain[ing]
the shortcomings it perceives so that the applicant is
properly notified and able to respond.” Hyatt, 492 F.3d at
1370. In other words, the PTO carries its procedural
burden of establishing a prima facie case when its rejection
satisfies 35 U.S.C. § 132, in “notify[ing] the applicant ...
[by] stating the reasons for [its] rejection, or objection or
requirement, together with such information and
references as may be useful in judging of the propriety of
continuing the prosecution of [the] application.” 35 U.S.C.
§ 132. That section “is violated when a rejection is so
uninformative that it prevents the applicant from
recognizing and seeking to counter the grounds for
rejection.” Chester v. Miller, 906 F.2d 1574, 1578
(Fed.Cir.1990).
In re Jung, 637 F.3d 1356, 1362 (Fed. Cir. 2011).
In this case, the Examiner states:
Dependent claims 10 and 21-36 when analyzed as a whole,
considering the additional elements individually and/or as
an ordered combination, are held to be patent ineligible
under 35 U.S.C. 101 because the additional recited
limitations fail to establish that the claims are directed to
an abstract idea without significantly more. These claims
fail to remedy the deficiencies of their parent claims
above, and therefore rejected for at least the same rationale
as applied to their parent claims above, and incorporated
herein.
Final Act. 9. We find that the above explanation of the rejection is sufficient
to inform Appellant of the Examiner’s position that the additional elements
of the dependent claims, whether considered individually or as part of an
ordered combination, (1) also recite an abstract method of organizing human
activity and/or mental processes, (2) do not integrate the abstract idea into a
Appeal 2020-005211
Application 15/339,973

22
practical application, and (3) do not amount to significantly more than the
abstract idea, for reasons similar to those discussed above with respect to
claim 1. Indeed, Appellant did “recogniz[e] and seek[] to counter” the
rejection for at least some of the dependent claims (i.e., claims 10, 22–24,
26, 27, 30–32, 34, and 35). We next address Appellant’s arguments with
respect to these claims.
ii. Claim 10
Claim 10 recites the method of claim 1, further comprising:
receiving, by the data processing system, a response to
the notification from the computing device associated with the
authorized user, the response indicating agreement or
disagreement by the authorized user with the recommended
modification; and
in response to the response indicating agreement by the
authorized user:
modifying, by the data processing system, the
aggregate medication listing data structure to implement
the recommended modification; and
sending, by the data processing system, a
modification notification output to a computing system
associated with a dispensing source that dispenses a
medication associated with the recommended
modification to the patient, the modification notification
output indicating, to the dispensing source, the
modification.
Appeal Br. 62 (Claims App.).
Appellant contends that these additional steps “are not directed to
certain methods of organizing human activity” involving the interaction of a
user and a patient. Appeal Br. 31–32. Appellant further contends that the
dependent limitations include a step that is “not just some abstract idea, but
rather a specific interaction of computing devices to implement an
improvement in the dispensing of medications by the dispensing source,”
Appeal 2020-005211
Application 15/339,973

23
which “clearly set[] forth a practical application of any alleged abstract
idea.” Id. at 31.
We are not persuaded. The additional limitations of claim 10 recite
using a computer to perform the steps of receiving a response from an
authorized user (e.g., the patient’s medical provider) indicating whether they
agree with a recommended modification to a patient’s aggregate medication
list, and notifying the dispensing source of the recommended modification if
the authorized user agree with the recommendation. For the same reasons
discussed above with respect to the limitations of claim 1, these additional
limitations also recite part of a method of organizing human activity and,
thus, an abstract idea. Neither do the additional limitations recite a practical
application, for the same reasons discussed above: the alleged improvement
is itself an ineligible abstract idea, rather than an improvement to the
functioning of the computer or a non-abstract improvement to another
technical field.
iii. Claim 22, 30
Appellant contends that claims 22 and 30 recite specific steps for
determining whether a medication is a valid or invalid duplicate, including:
performing a first determination as to whether the
medication is a potential duplicate of another medication
by determining whether there is a first instance of the
medication in the patient medical data having a first
source and at least one second instance of the medication,
or a medication in the same class of medication, in the
patient medical data having a second source different
from the first source. Then, specifically in response to
determining that the medication is a potential duplicate,
performing a second determination, based on the
application of evaluation rules implemented in logic of
the duplicate medication analysis engine, as to whether
Appeal 2020-005211
Application 15/339,973

24
the medication is a valid or invalid duplicate medication
by evaluating a temporal proximity of the first instance
and the second instance and differences in types and
geographical location of the first source and second
source. Thereafter, specifically in response to
determining that the medication is an invalid duplicate
medication, determining that the modification to the
aggregate medication listing data structure is to be
performed.
Appeal Br. 32. Appellant contends that these steps do not constitute “a
method of organizing human activity as alleged by the Final Office Action,”
but are rather “a specific computer process of the improved computer tool of
the claimed invention for determining valid/invalid medication duplicates
and modifying a medication list for a patient accordingly.” Id.
We are not persuaded for the same reasons discussed above with
respect to claim 1. We agree the additional limitations of claims 22 and 30
recite particular steps for determining whether a medication is a valid or
invalid duplicate; however, these steps are part of a method of organizing
human activity and/or mental processes (i.e., determining and evaluating)
and thus recite abstract ideas. The limitations do not improve the
functionality of the computer itself. Customedia Tech., 951 F.3d at 1365.
iv. Claims 23, 24, 31, 32
Appellant contends that
Claims 23-24 and 31-32 are dependent from claims 22
and 30, respectively, and provide further recitations of
the particular evaluation performed by the computer
operations to determine whether a medication is a valid
or invalid duplicate based on distance between locations
of sources and temporal duration. Again, these are further
defining specific computer operations for evaluating
medication listings specifically with regard to valid and
invalid duplicate medication instances. This is not some
Appeal 2020-005211
Application 15/339,973

25
abstract idea, but a specific computer operation. Thus,
claims 23-24 and 31-32, like claims 22 and 30, are patent
eligible and the Final Office Action has failed to establish
a prima facie case otherwise.
Appeal Br. 32–33. We are not persuaded for the same reasons
discussed above.
v. Claim 26, 34
Claims 26 and 34 depend indirectly from independent claims 1 and
11, respectively, and further recite the following limitations:
[ingesting/ingest], by the cognitive computing system,
the modified aggregate medication listing data structure and
patient electronic medical record from the patient registry;
[performing/perform], by the cognitive computing
system, a cognitive treatment recommendation operation based
on the modified aggregate medication listing data structure and
patient electronic medical record to generate a treatment
recommendation for the patient; and
[outputting/output], by the cognitive computing system,
the treatment recommendation for the patient to a computing
device associated with a medical practitioner.
Appeal Br. 67–68, 71 (Claims App.).
Appellant contends that
[i]ngestion is a process by which the cognitive computing
system . . . extracts features from . . . unstructured
information [from a corpus] to generate an in-memory
representation of the unstructured information that is
usable by the cognitive computing system to perform its
cognitive operations. Thus, claims 26 and 34 are clearly
not directed to any method of organizing human activity,
but rather to a process of improving the way in which a
cognitive computing system performs a cognitive
treatment recommendation operation based on a modified
aggregate medication listing data structure.
Id. at 33.
Appeal 2020-005211
Application 15/339,973

26
We are not persuaded because taking in data (“ingesting” data) is a
process that may be performed by a human being by, e.g., reviewing the
data, and reviewing data to generate patient treatment recommendations is
part of the process of medication reconciliation, which as we have discussed
above is a method of organizing human activity and an abstract idea. The
fact that the claims use a computer system (cognitive computing system) to
perform the additional steps in the dependent claims does not by itself render
the claim patent-eligible. See, e.g., In re Salwan, 681 Fed. App’x at 941;
Intellectual Ventures I LLC, 838 F.3d at 1314–15.
Furthermore, the recited steps of receiving and extracting data (i.e.,
“ingesting”), analyzing it (i.e., “performing a[n] . . . operation . . . to
generate a . . . recommendation”), and outputting the result are generic
computer functions. buySAFE, 765 F.3d at 1355 (receiving and sending
information over a network not inventive); Stanford, 991 F.3d 1245 at 1251–
1252 (Fed. Cir. 2021) (receiving, extracting, and storing data are well-
known, routine, and conventional functions of a regular computer). Neither
has Appellant cited to any persuasive evidence that the claims require the
recited cognitive computing system to perform any operations that are not
conventionally and routinely performed by a computer.
vi. Claims 27, 35
Appellant contends with respect to claims 27 and 35 that the
determination of whether to modify the aggregate medication listing data
structure, and the generation and comparison of various scores recited in
these claims for making such a determination, “clearly are not organizing
any human activity and are not operations that one would practically
Appeal 2020-005211
Application 15/339,973

27
perform mentally.” Appeal Br. 33–34. We are not persuaded for reasons
already discussed above.
In summary, the combinations of elements recited in claims 1, 10, 22–
24, 26, 27, 30–32, 34, and 35, do not amount to significantly more than the
judicial exceptions themselves, and under 35 U.S.C. § 101 the claimed
method is not eligible for patenting. As discussed above, Appellant has not
separately argued independent claims 11 and 20, and they fall with claim 1.
Appellant likewise fails to separately argue dependent claims 21, 25, 28, 29,
33, and 36, and they fall with claims 1 and 11.
B. Obviousness rejection over Miller and Boone (claims 1, 10, 11, 20,
21, 25, 26, 29, 33, 34)
1. Issue
The Examiner finds that Miller teaches all of the limitations of claims
1, 10, 11, 20, 21, 25, 26, 29, 33, and 34, except that Miller does not
explicitly teach “identify[ing] the medication related content within the
plurality of patient medical data at least by performing computer executed
natural language processing on the plurality of patient medical data to
extract portions of natural language text corresponding to medications
specified in the plurality of patient medical data.” Appeal Br. 60 (Claims
App.) (emphasis added); Final Act. 10–23. However, the Examiner finds
that Boone teaches this limitation. Final Act. 16.
The Examiner concludes that, at the time of the invention, it would
have been obvious to a skilled artisan to combine Miller and Boone to arrive
at the claimed invention, because “Miller teaches creating medication lists
. . . by processing patient data” while “Boone teaches using natural language
processing to process patient data,” and “combin[ing] these two concepts . . .
Appeal 2020-005211
Application 15/339,973

28
would allow for more accurate data” and merely involves “the simple
substitution of one known element for another [that] produces a predictable
result.” Final Act. 16–17.
Appellant contends that the combination of Miller and Boone does not
teach or suggest “the concept of valid/invalid duplicate medication
instances” or “the specific operations of the claimed invention to determine
valid/invalid duplicate medication instances.” Appeal Br. 37–39. Appellant
further contends that the combination of Miller and Boone does not teach or
suggest “the specific natural language processing recited in the . . . claims.”
Id. at 40.
2. Analysis
Claims 1, 11, 20, 25, 26, 33, 34
Unless otherwise noted, we adopt the Examiner’s findings of fact and
reasoning regarding the scope and content of the prior art (Final Act. 10–17;
Ans. 6–8) and agree with the Examiner that claim 1 is obvious over Miller
and Boone. Appellant does not separately argue claims 11, 20, 25, 26, 33,
and 34, and those claims fall with claim 1. We address Appellant's
arguments below.
i. Valid/Invalid duplicate medication
Appellant concedes that Miller teaches evaluating whether a
medication is a duplicate of another medication and, if so, removing the
duplicate. Appeal Br. 39. However, Appellant contends that the
combination of Miller and Boone does not teach or suggest “the concept of
valid/invalid duplicate medication instances.” Id. at 39; see also Reply Br.
14.
Appeal 2020-005211
Application 15/339,973

29
We are not persuaded. The claim defines a valid duplication as “a
duplicate that is determined should remain in the aggregate medication
listing data structure,” whereas “an invalid duplicate is a duplicate that is
determined should be removed from the aggregate medication listing data
structure.” Appeal Br. 61 (Claims App.). Miller teaches evaluating a
patient’s medication list to identify potentially adverse health outcome,
which may include “determining potential interactions between various
medications being used by the patient” and “identify[ing] medications which
may no longer be required due to duplication of the same or similar
medications.” See, e.g., Miller ¶ 50.
Miller further teaches that, once the potentially adverse health
outcomes have been identified and evaluated, a “health risk index or
medication therapy index” may be developed to indicate “the degree to
which the identified adverse health outcome may occur” based on factors
such as “number of medications, therapeutic classes of medications, number
of pharmacies, number of doctors, inferred disease(s), actual disease(s), age
and gender of patient, and laboratory data.” See, e.g., id. ¶¶ 51–52. Miller
teaches that the results of the evaluation may include potential solutions or
recommendations, such as replacing a medication or to consult a medical
provider. Id. ¶ 52. Finally, Miller teaches that, “based on the likelihood and
severity of any identified, potentially adverse health outcome,” its system
determines whether an active intervention, such as contacting the patient’s
medical provider, is necessary. Id. ¶ 54.
Based on the above disclosures, Miller’s system does not teach that all
identified duplicates should be removed from the patient’s medication list.
Instead, Miller teaches that, once a potentially adverse health outcome is
Appeal 2020-005211
Application 15/339,973

30
identified — such as medication that is no longer needed due to duplication
— a further evaluation is made as to the likelihood and severity of a
potentially adverse health outcome, before a recommendation and/or
intervention is made. Thus, contrary to Appellant’s contention, Miller
suggests “the concept of valid/invalid duplicate medication instances” (i.e.,
duplicate medication instances that the system determines should
remain/should be removed).
Appellant further contends that Miller does not teach or suggest “the
specific operations of the claimed invention to determine valid/invalid
duplicate medication instance.” Appeal Br. 39. In particular, Appellant
argues that in Miller
[t]here is no teaching or suggestion to specifically classify
medications specified in the aggregate medication listing
data structure into corresponding classes of medication
based on the related data types. There certainly is no
teaching or suggestion to specifically determine, for each
medication in a same class of medication in the
medication listing data structure, whether the medication
is a valid duplicate or an invalid duplicate medication
relative to one or more other medications in the same
class of medication based on an application of evaluation
rules implemented in logic of the duplicate medication
analysis engine to characteristics of the medication and
characteristics of the one or more other medications in
the same class, wherein a valid duplicate is a duplicate that
is determined should remain in the aggregate medication
listing data structure, and an invalid duplicate is a
duplicate that is determined should be removed from the
aggregate medication listing data structure.
Id. 39–40.
We are not persuaded. Miller teaches an “intelligent network system
for sharing medication information, medical information and medical
Appeal 2020-005211
Application 15/339,973

31
provider information,” which includes “analytical services, such as a merged
claim analytics database” that “may be used to accumulate, analyze or
download data relating to . . . weighted data or prescriptions for
medications,” resulting in a “patient file [that] may thereby potentially
include a complete, or near-complete, identification of the patient
medication therapy regimen, including medical conditions, medications and
information related thereto.” Miller ¶¶ 5, 19–21. Thus, Miller teaches the
“aggregate medication listing structure” recited in claim 1.
Furthermore, as discussed above, Miller teaches “data types” that may
be used to assess health risks posed by a patient’s medications, such as “the
name of the medication, potential side effects, potential interactions with
other medications, medical conditions the medication is meant to address,
the patient's duration of use and the patient’s utilization of a medication such
as a rate or amount of use, rate or amount of refills, medication dosage, etc.”
Miller ¶ 46; compare with Spec. ¶ 50 (describing “medication related data
types” as including, among others, medication type (e.g., anti-inflammatory
or pain reliever); date of service; institution (e.g., pharmacy, hospital,
primary care physician)). Thus, Miller teaches or suggests classifying a
patient’s medications into classes based on data types.
Finally, as also discussed above, Miller teaches identifying potentially
adverse health outcomes by, e.g., “identify[ing] medications which may no
longer be required due to duplication of the same or similar medications,”
and then evaluating the likelihood of an adverse health outcome occurring
based on factors such as, e.g., “number of medications, therapeutic classes of
medications, number of pharmacies, number of doctors, inferred disease(s),
actual disease(s), age and gender of patient, and laboratory data.” Miller
Appeal 2020-005211
Application 15/339,973

32
¶ 50–52. We find that this teaches or suggests determining whether a
medication should be removed (i.e., is or is not a valid duplicate) based on a
comparison with medication in the same class (i.e., “same or similar
medications”) based on characteristics of the potentially duplicative
medications (i.e., “characteristics of the medication and characteristics of the
one or more other medications in the same class”).
ii. Natural language processing
With respect to the limitation of analyzing patient medical data to
identify medication related content by “performing computer executed
natural language processing . . . to extract portions of the natural language
text corresponding to medications,” Appellant contends that
Boone is directed to a mechanism for utilizing natural
language medical records. What Boone actually teaches
. . . is that medical information text is placed into
predetermined medical record templates where text
pertinent to a section of the template is entered in that
section as natural language, and information about the
sections is used to extract the natural language from the
template. This is not natural language processing, but
rather just entry of text into a template and retrieval of text
from the template based on what sections of the template
the text was entered into. This certainly is not the specific
natural language processing recited in the present claims,
i.e. natural language processing performed on a plurality
of patient medical data to extract portions of natural
language text corresponding to medications specified in
the plurality of patient medical data.
Appeal Br. 40–41; see also Reply Br. 14–15.
We are not persuaded. As Appellant acknowledges, Boon teaches
“extracting elements of medical data from the [natural language] text,”
which is entered into a medical record template that is then stored as a
Appeal 2020-005211
Application 15/339,973

33
completed medical record, wherein “information about which section of the
template the text was entered into is used to extract elements of medical data
from the text.” Boon ¶ 25. We find this to fall within the broadest
reasonable interpretation of “performing computer executed natural
language processing . . . to extract portions of the natural language text
corresponding to” medical data, such as information regarding medications.
Appellant has provided only conclusory statements, and no persuasive
evidence or argument, why the term should be more narrowly construed. Cf.
In re Self, 671 F.2d 1344, 1348 (CCPA 1982) (“[A]ppellant’s arguments fail
from the outset because . . . they are not based on limitations appearing in
the claims.”).
Claim 10
With respect to dependent claim 10, Appellant further contends that
the alleged combination of Miller and Boone does not
teach or render obvious the additional specific features of
receiving, by the data processing system, a response to the
notification from the computing device associated with
the authorized user, the response indicating agreement or
disagreement by the authorized user with the
recommended modification; and in response to the
response indicating agreement by the authorized user:
modifying, by the data processing system, the aggregate
medication listing data structure to implement the
recommended modification; and sending, by the data
processing system, a modification notification output to a
computing system associated with a dispensing source
that dispenses a medication associated with the
recommended modification to the patient, the
modification notification output indicating, to the
dispensing source, the modification.
Appeal Br. 43–44; see also Reply Br. 15–16.
Appeal 2020-005211
Application 15/339,973

34
We are not persuaded. Miller teaches that medical providers and
other authorized users may be provided with the results of the health risk
assessment and that potential solutions and/or recommendations that may be
made by its system with respect to identified, potentially adverse health
outcome include “replacing a medication” and/or “consult medical
provider.” Miller ¶ 52. Miller also teaches, depending on the likelihood and
severity of identified, potentially adverse health outcome, intervening in the
patient’s medication therapy regimen by contacting a patient’s physician.
Id. ¶ 62.
In short, Miller suggests sending the results of its health risk
assessment and proposed solutions to authorized users such as a patient’s
physician, and further teaches consulting with a medical provider regarding
identified, potential adverse health outcomes. Based on these disclosures in
Miller, we find that it would have been an obvious step for Miller’s
intelligent network system to receive a response from the authorized user
(e.g., a medical provider it consulted), agreeing or disagreeing with the
system’s proposed solution (e.g., replacing a particular medication)
regarding an identified, potentially adverse health outcome. We find that it
would have likewise been obvious, based on these disclosures, for the
system to modify the patient’s medication list once the authorized user has
agreed to the modification, as a skilled artisan would have understood that
confirming proposed solutions to potentially adverse health outcomes (e.g.,
by modifying the patient’s medication list) to be the purpose of consulting a
medical provider. In this regard, we note that the obviousness analysis “can
take account of the inferences and creative steps that a person of ordinary
skill in the art would employ.” KSR Int’l Co. v. Teleflex Inc., 550 U.S. 398,
Appeal 2020-005211
Application 15/339,973

35
418 (2007). “A person of ordinary skill is also a person of ordinary
creativity, not an automaton.” Id. at 421.
Finally, Miller teaches updating the patient’s medication therapy
regimen and also teaches “prompt[ing] a user, such as the patient’s
pharmacist or a pharmacist at a pharmaceutical care center 20, to intervene
in the patient’s medication therapy regimen to execute the new medication
therapy regimen.” Id. ¶¶ 93–96. Thus, Miller teaches sending a notification
of modifications to a patient’s medication therapy regimen (i.e., the “new”
medication therapy regimen) to a dispensing source (i.e., the pharmacist that
execute the new medication therapy regimen).
Accordingly, we affirm the Examiner’s rejection of claim 10 as
obvious over Miller and Boone.
Claims 21, 29
Appellant contends that,
with regard to claims 21 and 29, the alleged combination
of Miller and Boone fails to teach or suggest determining
whether the medication is a valid duplicate or an invalid
duplicate comprises determining whether the medication
is one of a replacement instance of the medication
relative to another instance of the medication in the same
class of medication, a short term or emergency
administering of the medication at a medical facility, or a
refill of an expired prescription.
Appeal Br. 44; see also Reply Br. 16–17.
We are not persuaded. Miller teaches that “duplication of a
medication may result in an increased cost without any additional benefit, as
well as a potential medical disadvantage.” Miller ¶ 3. Miller teaches that,
“[i]n some cases, medications may be replaced or combined by prescribing
an alternative medication that has an improved medical effect and/or may
Appeal 2020-005211
Application 15/339,973

36
also result in decreased cost to the patient.” Id. Miller further teaches
creating a new medication regiment including removing and/or replacing
one or more medications with a new medications, explaining that,
[f]or example, a medication may be removed if it is
duplicative of another medication, if the medication is no
longer required (e.g., the associated medical condition no
longer exists), if the removal of the medication will
improve the medical effect of the other medications
without adversely affecting the patient, if the removal will
reduce or eliminate an undesired side effect or if the
medication was without indication. A medication may be
replaced with another medication, including replacing a
brand name medication with a lower cost, equivalent
generic medication, if available.
Id. ¶ 94. Thus, Miller teaches determining that a duplicate medication (e.g.,
a brand name medication) should be removed (i.e., determining the
medication to be an invalid duplicate) where the duplicate medication has
been replaced, while determining that the replacement medication (e.g., a
generic medication) should remain in the medication list (i.e., determining
the medication to be a valid duplicate).
Accordingly, we affirm the Examiner’s rejection of claims 21 and 29
over Miller and Boone.
C. Obviousness rejection over Miller, Boone, Biernacki (claims 22,
23, 27, 28, 30, 31, 35, 36)
1. Issue
As discussed above, the Examiner finds that Miller teaches all of the
limitations of the claims from which claims 22, 23, 27, 28, 30, 31, 35, and
36 depend, except that Miller does not teach the limitation regarding natural
language processing,” which the Examiner finds to be disclosed in Boone.
Final Act. 10–23. The Examiner further finds that Miller teaches almost all
Appeal 2020-005211
Application 15/339,973

37
of the additional limitations in claims 22, 23, 30, and 31, Final Act. 25–27
and 34, except neither Miller nor Boone teaches, “in response to determining
that the medication is a potential duplicate, performing a second
determination . . . as to whether the medication is a valid or invalid duplicate
medication by evaluating a temporal proximity of the first instance and the
second instance [of the medication in the patient medical data] and
differences in types and geographical location of the first source and second
source [of the medication],” as required by the claims. Appeal Br. 66, 70
(Claims App.); Final Act. 25, 34.
However, the Examiner finds that Biernacki teaches these limitations.
Id. at 26, 34. The Examiner further cites Biernacki as teaching the
additional limitations in dependent claims 27, 28, 35, and 36, except the
Examiner cites Miller for teaching the limitation regarding “the aggregate
medication listing data structure [being] automatically modified to remove
the medication.” Final Act. 27–34.
The Examiner concludes that, at the time of the invention, it would
have been obvious to a skilled artisan to combine the teachings of Miller,
Boone, and Biernacki to arrive at the inventions of claims 22, 23, 27, 28, 30,
31, 35, and 36. In particular, the Examiner finds that “[t]he combination of
Miller in view of Boone discloses a system that can use natural language
processing to determine a list of medications and remove invalid/valid
duplicates to create an aggregated medication list.” Final Act. 25–26, 32,
33–34. The Examiner concludes that, as for claims 22, 23, 30, and 31, it
would have been obvious to combine this system with Biernacki’s teaching,
in the context of “a medication risk listing system,” of “evaluating a
temporal proximity between instances and differences in . . . geographic
Appeal 2020-005211
Application 15/339,973

38
locations of where the drug is manufactured,” because “doing so would
create more data regarding . . . where the medication is produced and the
distribution to the patient.” Id. at 26.
Similarly, the Examiner concludes that it would have been obvious to
a skilled artisan to combine the system suggested by Miller and Boone, with
Biernacki’s teaching of “using tiered risk stratification scores from different
risk value data sets” in the context of “a medication risk listing system,” to
arrive at the invention of claims 27, 28, 35, and 36, because “doing so would
create more data regarding the medication’s risks and allow for better patient
outcomes.” Id. at 32, 33–34.
The Examiner notes that the combinations in all of the above cases
would be further obvious because they would be “merely a combination of
old elements . . . perform[ing] the same function as [they] did separately,”
with a predictable result. Final Act. 26, 32, 34.
Appellant contends that the dependent claims subject to this rejection
are non-obvious for the same reasons discussed above with respect to the
rejection over Miller and Boone. Appeal Br. 46. Appellant further contends
that the combination of Miller, Boone, and Biernacki does not disclose
additional limitations recited in these dependent claims. Id. at 46–57.
The issue with respect to this rejection is whether preponderance of
the evidence supports the Examiner’s conclusion that claims 2, 23, 27, 28,
30, 31, 35, and 36 are obvious over Miller, Boone, and Biernacki.
Appeal 2020-005211
Application 15/339,973

39
2. Analysis
Claims 22, 23, 30, 31
Claims 22 and 30 depend from claims 1 and 11 and further recite the
following step, among others:
in response to determining that the medication is a potential
duplicate, performing a second determination, based on the
application of evaluation rules implemented in logic of the
duplicate medication analysis engine, as to whether the
medication is a valid or invalid duplicate medication by
evaluating a temporal proximity of the first instance and the
second instance and differences in types and geographical
location of the first source and second source.
Appeal Br. 66, 70 (Claims App.). Claims 23 and 31 depend from claims 22
and 30 and further recite:
Wherein performing the second determination . . .
comprises determining that the medication is a valid
duplicate medication in response to at least one of a
location of the first source or second source being within a
predetermined distance of a known location associated
with the patient.
Id. at 67, 70.
Appellant contends that, although Miller contains various teachings
regarding “identifying inappropriate medications in . . . patient’s medication
information to identify potentially adverse health outcomes, and
interventions in the person’s medication therapy regimen to prevent the
adverse health outcome,” Miller does not teach “any specific determination
of the differences [between the first and second instance of medication]
enumerated” in claims 23 and 31, “or that such determinations are used to
identify valid/invalid duplicate medication instances.” Appeal Br. 48–49.
Appellant similarly contends that, while Biernacki teaches “documenting
Appeal 2020-005211
Application 15/339,973

40
actions and responses taken that are associated with risk scores ‘over time’”
and also teaches that “generic drugs can have unique national drug codes
and [that] differences exist based on various drug manufacturing locations,”
these disclosures do not teach or suggest the claimed operations of
determining “temporal proximity” or “difference in a geographical location
of sources of medication instances.” Id. at 49.
With respect to claims 23 and 31, Appellant further contends that
the alleged combination of references fails to teach or
render obvious the feature that performing the second
determination based on the application of evaluation
rules implemented in logic of the duplicate medication
analysis engine comprises determining that the
medication is a valid duplicate medication in response to
at least one of a location of the first source or second
source being within a predetermined distance of a
known location associated with the patient.
Appeal Br. 50.
We agree with Appellant that the Examiner has not established a
prima facie case that claims 22, 23, 30, and 31 are obvious over the
combination of Miller, Boone, and Biernacki.
As to claims 22 and 30, the Examiner concedes that Miller and Boone
does not explicitly teach determining whether a medication is a valid or
invalid duplicate by “evaluating a temporal proximity of the first instance
and the second instance and differences in types and geographical location
of the first source and second source.” Final Act. 25. However, the
Examiner finds that the limitations regarding temporal proximity are
suggested by Biernacki’s teaching of “documenting action(s) and associated
responses taken that can be associated with risk scores over time and data-
Appeal 2020-005211
Application 15/339,973

41
mined for correlations where the strongest association between intervention
and risk are observed.” Id. (citing Biernacki ¶ 115).10
We are not persuaded. In paragraph 115, Biernacki teaches
documenting interventions (i.e., “action(s) and associated responses taken”)
and further teaches that these interventions can be associated with risk scores
over time so as to determine “correlations where the strongest association
between intervention and risk are observed.” Biernacki ¶ 115. While this
disclosure does discuss the concept of a time interval (i.e., “over time”), the
Examiner has not persuasively explained how the time interval disclosed in
Biernacki (i.e., risk scores over time) would suggest, to an ordinarily skilled
artisan, to “evaluat[e] a temporal proximity of the first instance and the
second instance [of medication],” or to use such evaluation to determine
whether the medication is a valid or invalid duplicate.11

10 The Examiner finds that the limitations relating to differences in types and
geographical location of the sources of the medication are suggested by
Biernacki’s teaching that the “precision of information used in assessing risk
factors associated with medication regimen” is improved by “obtaining data
from various levels of hierarchies of drug information, including particular
coding schemas” that may distinguish drugs based on, e.g., “drug
manufacturing locations, manufacturers and vendors.” Final Act. 25 (citing
Biernacki ¶ 62).
11 As discussed above, the Examiner concludes that it would have been
obvious to combine the system suggested by Miller and Boone with
Biernacki’s teaching, in the context of “a medication risk listing system,” of
“evaluating a temporal proximity between instances and differences in . . .
geographic locations of where the drug is manufactured,” because “doing so
would create more data regarding . . . where the medication is produced and
the distribution to the patient.” Id. at 26. Even assuming Biernacki suggests
evaluating a temporal proximity between instances of medication, and
assuming that a skilled artisan would have been motivated to incorporate
such information into the medication therapy management system suggested
Appeal 2020-005211
Application 15/339,973

42
Accordingly, we reverse the Examiner’s rejection of claims 22 and 30
over Miller, Boone, and Biernacki. Claims 23 and 31 depend from claims
22 and 30. We reverse the Examiner’s rejection of these claims for the same
reasons. In re Fine, 837 F.2d 1071, 1076 (Fed. Cir. 1988) (“Dependent
claims are nonobvious under section 103 if the independent claims from
which they depend are nonobvious.”).12
Claims 27, 35
Claims 27 and 35 depend from claims 1 and 11, respectively, and
further recite generating a duplicate medication score, a contraindication
score, a medication interaction score, a cohort analysis score, and an
aggregate score generated from combination of the scores, comparing the
aggregate score to one or more threshold scores, and determining whether

by Miller and Boone, the Examiner still has not explained why a skilled
artisan would have been motivated to use the information in determining
whether a medication is a valid duplicate. “[R]ejections on obviousness
grounds cannot be sustained by mere conclusory statements; instead, there
must be some articulated reasoning with some rational underpinning to
support the legal conclusion of obviousness.” KSR, 550 U.S. at 418.
12 We note that, with respect to the additional limitations of claims 23 and
31, the Examiner relies on Miller’s disclosure of a medication therapy
management method that may replace a duplicate drug and further relies on
Miller’s disclosure that information regarding a patient’s local pharmacist
may be requested during the patient enrollment process and that the
pharmacist may review all of a patient’s medications. Final Act. 26–27.
The Examiner provides no explanation, however, as to how these two
separate disclosures would provide a skilled artisan with reason to
“determin[e] that the medication is a valid duplicate medication in response
to at least one of a location of the first source or second source being within
a predetermined distance of a known location associated with the patient,” as
required by Appellant’s claimed invention.
Appeal 2020-005211
Application 15/339,973

43
the medication listing data structure is to be modified based on the results of
the comparison. Appeal Br. 68–69 (Claims App.).
Appellant acknowledges that Biernacki “general teach[es] risk
scoring”; however, Appellant contends that Biernacki does not “teach or
render obvious the specific scorings and generation of aggregate score
recited in the present claims, or the comparison of the aggregate score to one
or more thresholds to determine a medication listing modification.” Appeal
Br. 56.
We are not persuaded. Appellant concedes that Biernacki teaches the
general concept of risk scoring. Appeal Br. 56; see also Bernacki ¶ 78. The
combination of Miller and Biernacki further teaches that factors that may be
taken into account in assessing the health risk of, and optimizing, a patient’s
medication therapy regimen include, among other things, whether
medications are duplicated, see, e.g., Miller ¶ 50; whether a medication may
be harmful when specific factors regarding a patient’s situation, such as age,
are taken into account (i.e., whether a medication is contraindicated), see,
e.g., Biernacki ¶ 76, Miller ¶¶ 42, 50, 97–98; whether two medications may
interact with each other, see, e.g., Biernacki ¶ 71, Miller ¶ 50; and treatment
guidelines and/or industry recognized standards relating to a patient’s
medical condition (i.e., patient cohorting), see, e.g., Miller ¶¶ 82, 106–108.
We find that the combination of these disclosures renders obvious the
limitations in claims 27 and 35 regarding the generation of a “duplicate
medication score,” a “contraindication score,” a “medication interaction
score,” and a “cohort analysis score.”
Biernacki also teaches embodiments of its invention that “provide for
calculating risk based on a comprehensive assessment and annotation of the
Appeal 2020-005211
Application 15/339,973

44
data sets,” as well as for “direct association of recommended interventions
and responses to specific risk ‘triggers’ . . . and data elements.” Biernacki
¶¶ 25–26; see also id. ¶ 28 (describing aggregating risk scores), ¶ 78
(describing “[f]inal risk scores for a given source or channel” that are “the
aggregate of individual scores, summed and optionally weighted or
normalized”), ¶¶ 107–108 (explaining that “risk scores can be automatically
and dynamically calculated for each distinct data source via the application
of a rules engine to a variety of factors,” wherein “[r]ules can be applied by
assessing the presence of a number of combined factors, and based on the
combination of actors, determining an individual risk score . . . for each
unique combination of factors”). Miller likewise teaches assigning patients
a health risk index or medication therapy index score based on the severity
and likelihood of identified, potentially adverse outcomes and
“determin[ing] whether an active intervention is necessary.” Miller ¶¶ 53–
54. We find that the combination of these teachings suggest the limitations
in claims 27 and 35 regarding combining various risk scores discussed above
to generate an aggregate score for a medication, and to determine whether to
modify patient’s medication list (i.e., to intervene) based on comparing the
aggregate score to one or more thresholds.
Accordingly, we affirm the Examiner’s rejection of claims 27 and 35
over Miller, Boone, and Biernacki.
Claims 28, 36
Appellant contends that,
[w]ith regard to claims 28 and 36, the alleged combination
of references fails to teach or render obvious the additional
features of, in response to the aggregate score for the
medication being equal to or greater than a
Appeal 2020-005211
Application 15/339,973

45
predetermined threshold score in the one or more
threshold scores, the aggregate medication listing data
structure is automatically modified to remove the
medication from the aggregate medication listing data
structure.
Appeal Br. 56–57.
In particular, Appellant contends that, although Miller teaches
“identifying inappropriate medications in a patient’s medication therapy
regimen” and “removal of medications that are determined to be duplicates,”
Miller does not teach or suggest the claim limitations regarding “aggregate
score” or the automatic removal of a medication from a medication listing
when the aggregate score is equal or greater than a threshold. Appeal Br. 57.
Appellant contends that Biernacki’s teaching of “‘aggregation’ of data sets
. . . and the assessing of risk factors associated with a medication regimen”
also does not teach or suggest the above limitations. Id.
We are not persuaded. As discussed above, Biernacki teaches
aggregating risk scores, as well as associating recommended interventions to
specific risk triggers. Biernacki ¶¶ 25–26, 28, 78, 107–108, and Miller
similarly teaches assigning patients a health risk index or medication therapy
index score based on the severity and likelihood of identified, potentially
adverse outcomes and “determin[ing] whether an active intervention is
necessary.” Miller ¶¶ 53–54. Miller also teaches that, where the level of
risk associated with particular medications in the patient’s medication
therapy regimen warrants intervention, “the inappropriate medications
routine . . . may proceed to establish a new medication therapy regimen for
the patient,” which may involve “removing the inappropriate medication,
and/or replacing the inappropriate medication with a new medication having
Appeal 2020-005211
Application 15/339,973

46
a comparable medical effect without the attendant adverse health outcome.”
Id. ¶¶ 101–102.
We find that the combination of these disclosures—i.e., Biernacki’s
teaching regarding aggregating risk scores and Miller’s similar teaching
regarding a “health risk index or medication therapy index score,” the prior
art’s teachings regarding evaluation of the health risks for determining
necessity of intervention, and Miller’s teaching that intervention may
include establishment of a new medication therapy involving removal or
replacement of inappropriate medication—renders obvious the additional
limitations of claims 28 and 36. In this regard, we note that “[t]he
combination of familiar elements according to known methods is likely to be
obvious when it does no more than yield predictable results.” KSR, 550 U.S.
at 416.
Accordingly, we affirm the Examiner’s rejection of claims 28 and 36
over Miller, Boone, and Biernacki.
D. Obviousness rejection over Miller, Boone, Biernacki, Arshad
(claims 24, 32)
1. Issue
Claims 24 and 32 depend from claims 22 and 30, respectively and
further requires, in response to determining that a medication is a potential
duplicate, “determining that the medication is a valid duplicate medication in
response to the first source [of the medication] being within a predetermined
distance of the second source and a temporal duration between the first
instance [of the medication in the patient medical data] being within a
predetermined temporal period of the second instance based on the temporal
proximity.” Appeal Br. 67, 70–71 (Claims App.).
Appeal 2020-005211
Application 15/339,973

47
As discussed above, the Examiner finds that the combination of
Miller, Boone, and Biernacki teaches all of the limitations of claims 22 and
30. Final Act. 10–23, 25–27, 34. The Examiner finds that the combination
of Miller, Boone, and Biernacki does not teach the additional claim
limitations of claims 24 and 32. However, the Examiner finds that Arshad
teaches these limitations. Id. at 34–35.
The Examiner concludes that it would have been obvious to a skilled
artisan, at the time of the invention, to combine the teachings in Arshad to
show patient “valid duplicated medication information . . . in a
predetermined temporal period,” with the system suggested by the
combination of Miller, Boone, and Biernacki—i.e., “a system that can use
natural language processing to determine a list of medications and remove
invalid/valid duplicates to create an aggregated medication list using tiered
risk stratification scores from different risk value data sets”— to arrive at the
inventions of claims 24 and 32. Final Act. 35–36. The Examiner concludes
that such a combination would have been obvious “because doing so would
create more data regarding the medication’s risks and allow for the patient to
make an educated choice on the medication equivalents that are available to
them,” and because “the claimed invention is merely a combination of old
elements . . . perform[ing] the same function as [they] did separately,” with
predictable results. Final Act. 36.
Appellant contends that claims 24 and 32 are non-obvious for the
same reasons as discussed above with respect to the rejection of claims 22
and 30 over Miller, Boone, and Biernacki. Appeal Br. 58. Appellant also
contends that Arshad does not teach the additional limitations recited in
claims 24 and 32. Id. at 58–59.
Appeal 2020-005211
Application 15/339,973

48
The issue with respect to this rejection is whether preponderance of
the evidence supports the Examiner’s conclusion that claims 24 and 32 are
obvious over Miller, Boone, Biernacki, and Arshad.
2. Analysis
We agree with Appellant that the Examiner has not established a
prima facie case that claims 24 and 32 are obvious over Miller, Boone,
Biernacki, and Arshad.
In particular, claims 24 and 32 recite the additional limitation that
determining whether the medication is a valid or invalid duplicate further
comprises “determining that the medication is a valid duplicate medication
in response to the first source [of the medication] being within a
predetermined distance of the second source and temporal duration between
the first instance [of the medication] being within a predetermined temporal
period of the second instance based on the temporal proximity.” Appeal Br.
67, 70–71 (Claims App.). The Examiner asserts that Arshad’s paragraph
243 and accompanying Figure 17 teach the additional limitation of claims 24
and 32. Final Act. 34–36; Ans. 10.
Arshad teaches a telemedicine system. The cited disclosures describe
a user interface that provides an “estimated wait time for the encounter with
the healthcare provider” and further comprises “advertisement information
which may be directed to the patient,” such as advertisement for treatment or
medication from a third party for a patient’s medical condition (e.g.,
diabetes), where knowledge of the patient’s medical condition is derived
through analysis of patient information. Arshad ¶ 243. Arshad further
teaches that “[t]he duration of the advertisement is determined so that the
advertisement fits within a time period . . . of the wait notification” and that
Appeal 2020-005211
Application 15/339,973

49
the user may be provided with the option to “receive information about
and/or obtain a free sample of the directed medication.” Id.
As we understand it, the Examiner asserts that the patient information
from which Arshad’s system derives knowledge of the patient’s medical
condition reads on the “first instance” of the medication. Final Act. 35. The
Examiner asserts that the medication being advertised in Arshad’s system
reads on the “second instance” of the medication having a different source
than the first instance of the medication. Id. Finally, the Examiner appears
to assert that the temporal duration between the first instance of the
medication is within a predetermined temporal period of the second instance
because the advertisement is taught to fit with a prescribed period (i.e.,
within the estimated wait time). Id.
We are not persuaded. The claims require the first and second
instances of the medication to be in the patient medical data, and there is no
teaching in Arshad that the advertised medication is in the patient data
during the period of the advertisement. Neither has the Examiner provided a
persuasive construction of “temporal duration” between two potentially
duplicate medications that would encompass the duration of the
advertisement described in Arshad.13

13 The Examiner appears to construe these claims as requiring “having valid
duplicated medication information . . . shown to the patient in a
predetermined temporal period.” Final Act. 36. While we agree that the
PTO gives claims their broadest reasonable interpretation during
examination, the claims must nevertheless be interpreted as they would be
read by a skilled artisan and consistent with the Specification. In re
Cortright, 165 F.3d 1353, 1358, 49 USPQ2d 1464, 1467 (Fed. Cir. 1999); In
re Hyatt, 211 F.3d 1367, 1372 (Fed. Cir. 2000). Neither the claim language
itself nor the Specification supports the above interpretation of the claims.
Appeal 2020-005211
Application 15/339,973

50
Finally, the Examiner asserts that it would have been obvious to a
skilled artisan to combine the four cited references “because doing so would
create more data regarding the medication’s risks and allow for the patient to
make an educated choice on the medication equivalents that are available to
them,” and because the claims are merely the combination of familiar
elements according to known methods that yields predictable results. Final
Act. 35–36. Nevertheless, the Examiner has not provided articulated
reasoning as to why a skilled artisan would have incorporated Arshad’s
teaching regarding an medication advertisement having a predetermined
duration into the system suggested by the combination of Miller, Boone, and
Biernacki in the manner recited in the claim (i.e., as a factor in determining
whether a duplicate medication is valid or invalid). KSR, 550 U.S. at 418.
Accordingly, we reverse the Examiner’s rejection of claims 24 and 32
as obvious over Miller, Boone, Biernacki, and Arshad.
E. Rejection under 35 U.S.C. § 112 for lack of written description
(claims 22–24, 27, 28, 30–32, 35, 36)
1. Issue
Claims 27, 28, 35, 36
Claims 27 and 35 are dependent claims wherein “determining whether
a modification to the aggregate medication listing data structure [is to be
performed] comprises, for a medication . . .: generating . . . a duplicate
medication score,” “a contraindication score,” “a medication interaction
score,” and “a cohort analysis score”; combining the recited scores to
generate “an aggregate score” for the medication; and “comparing the
aggregate score for the medication to one or more threshold scores.”
Appeal Br. 68–69, 71–72. Claims 28 and 36 depend from claims 27 and 35
Appeal 2020-005211
Application 15/339,973

51
and further recite “wherein, in response to the aggregate score for the
medication being equal to or greater than a predetermined threshold score in
the one or more threshold scores, the aggregate medication listing data
structure is automatically modified to remove the medication from the
aggregate medication listing data structure.” Id. at 69, 72–73 (emphasis
added).
The Examiner finds that the Specification “fails to adequately disclose
generating a duplicate medication score, a contraindication score, a
medication interaction score, a cohort analysis score, an aggregate
score, and comparing the aggregate score for the medication to one or
more threshold score, and a predetermined threshold score,” because
“the specification fails to portray how the scores are generated or how the
scores are compared.” Final Act. 3.
Claims 23, 24, 31, 32
Claims 23 and 31 are dependent claims that further recite a limitation
of “determining that the medication is a valid duplicate medication in
response to at least one of a location of the first source or second source
being within a predetermined distance of a known location associated with
the patient.” Appeal Br. 67, 70 (Claims App.) (emphasis added).
Claims 24 and 32 are dependent claims that further recite a limitation
of “determining that the medication is a valid duplicate medication in
response to the first source being within a predetermined distance of the
second source and a temporal duration between the first instance being
within a predetermined temporal period of the second instance based on the
temporal proximity.” Appeal Br. 67, 70–71 (Claims App.) (emphasis
added).
Appeal 2020-005211
Application 15/339,973

52
The Examiner finds that “there is not a predetermined distance recited
anywhere in the specification that would adequately describe what a
predetermined distance is indicative of.” Ans. 5.
Claims 22, 30
Claims 22 and 30 are dependent claims that further recite a limitation
of “performing a . . . determination . . . as to whether the medication is a
valid or invalid duplicate medication by evaluating a temporal proximity of
the first instance and the second instance and differences in types and
geographical location of the first source and second source.” Appeal Br. 66,
70 (Claims App.).
The Examiner finds that “the specification, as originally filed, fails to
adequately disclose evaluating a temporal proximity of the first instance
and the second instance and differences in types and geographical
location of the first source and second source” because the Specification
“fails to portray how the temporal proximity is evaluated.” Final Act. 3.
2. Analysis
To satisfy the written description requirement, the Specification must
describe the claimed invention in sufficient detail such that a skilled artisan
can reasonably conclude the inventor possessed the claimed invention. Vas-
Cath, Inc. v. Mahurkar, 935 F.2d 1555, 1562–63 (Fed. Cir. 1991). “And
while the description requirement does not demand any particular form of
disclosure, or that the specification recite the claimed invention in haec
verba, a description that merely renders the invention obvious does not
satisfy the requirement.” Ariad Pharm., Inc. v. Eli Lilly & Co., 598 F.3d
1336, 1352 (Fed. Cir. 2010) (citation omitted).
Appeal 2020-005211
Application 15/339,973

53
On January 7, 2019, the Director of the USPTO issued specific
guidance on “Examining Computer-Implemented Functional Claim
Limitations for Compliance With 35 U.S.C. 112 (“2019 § 112 Guidance”).
84 Fed. Reg. 57 (Jan. 7, 2019). The 2019 § 112 Guidance emphasizes that,
“[e]ven if a claim is not construed as a means-plus-function limitation under
35 U.S.C. 112(f), computer-implemented functional claim language must
still be evaluated for sufficient disclosure under the written description and
enablement requirements of 35 U.S.C. 112(a).” 84 Fed. Reg. at 61. For
example, “a specification must describe the claimed invention in sufficient
detail (e.g., by disclosure of an algorithm) to establish that the applicant had
possession of the claimed invention as of the application filing date.” Id.
The 2019 § 112 Guidance notes that “[t]he written description requirement is
not met if the specification merely describes a ‘desired result.’” Id. (citing
Vasudevan Software, Inc. v. MicroStrategy, Inc., 782 F.3d 671, 682 (Fed.
Cir. 2015)).
In the context of evaluating the adequacy of the written description for
“computer-implemented, software-related claims,” therefore, we determine
whether the specification discloses the computer and the
algorithm(s) that achieve the claimed function in sufficient
detail that one of ordinary skill in the art can reasonably
conclude that the inventor possessed the claimed subject
matter at the time of filing. An algorithm is defined, for
example, as “a finite sequence of steps for solving a logical
or mathematical problem or performing a task.” Microsoft
Computer Dictionary (5th ed., 2002). Applicant may
“express that algorithm in any understandable terms
including as a mathematical formula, in prose, or as a flow
chart, or in any other manner that provides sufficient
structure.” Finisar [Corp. v. DirecTV Grp., Inc.], 523 F.3d
[1323,] 1340 [(Fed. Cir. 2008)] (internal citation omitted).
Appeal 2020-005211
Application 15/339,973

54
It is not enough that one skilled in the art could
theoretically write a program to achieve the claimed
function, rather the specification itself must explain how
the claimed function is achieved to demonstrate that the
applicant had possession of it. See, e.g., Vasudevan, 782
F.3d at 682–83.
84 Fed. Reg. at 61–62.
For the reasons discussed below, we agree with the Examiner that
claims 22–24, 27, 28, 30–32, 35, and 36 are invalid for lack of adequate
written description.
Claims 27, 28, 35, 36
We agree with the Examiner that the Specification does not
adequately describe claims 27, 28, 35, and 36. Claims 27 and 35 recite
determining whether a modification is to be performed by generating various
scores (a duplicate medication score, a contraindication score, a medication
interaction score, and a cohort analysis score), combing the scores to
generate an aggregate score, comparing the aggregate score to one or more
threshold scores, and determining whether to modify the medication listing
based on the results of the comparison. Claims 28 and 36 depend from
claims 27 and 35 and further recite automatically modifying the aggregate
medication listing to remove a medication where the aggregate score is equal
to or greater than the predetermined threshold score.
Appellant contends that “the Examiner has admitted that the claimed
invention is enabled and thus, there is a strong presumption that the
description is . . . adequate.” Appeal Br. 11. Appellant contends that this is
particularly the case because the limitations found to be inadequately
described are “features of determining values, e.g., distance, temporal
proximity, scores, and thresholds, and comparing specific ones of these
Appeal 2020-005211
Application 15/339,973

55
numerical values, and especially given the high level of skill in the art and
the predictability of computer based inventions.” Id. Appellant contends
that, “for the Examiner’s position to have merit, one would have to take the
position that . . . a person skilled in the art would not know how to calculate
the values recited in the present claims given the present specification” and
“must be given explicitly step-by-step instructions on how to calculate a
numerical value and compare that numerical value to other numerical
values.” Id.; see also Reply Br. 2, 4.
We are not persuaded. As an initial matter, and as Appellant
acknowledges but then largely brushes aside, Appeal Br. 10, enablement and
written description are separate and distinct requirements of patentability.
Ariad, 598 F.3d at 1352–1353 (explaining that the written description
requirement “plays a vital role in curtailing claims that do not require undue
experimentation to make and use . . . but that have not been invented” and
further “ensures that when a patent claims a genus by its function or result,
the specification recites sufficient materials to accomplish that function”).
Furthermore, contrary to Appellant’s argument, our reviewing court has
made it clear that the written description requirement is not satisfied merely
because the disclosure renders the claimed subject matter obvious to a
skilled artisan. See ICU Med, Inc. v. Alaris Med. Sys., Inc., 558 F.3d 1368,
1377 (Fed. Cir. 2009); see also Regents of the Univ. of Cal. v. Eli Lilly &
Co., 119 F.3d 1559, 1566-67 (Fed. Cir. 1997) (explaining that “an applicant
complies with the written description requirement by describing the
invention, with all its claimed limitations, not that which makes it obvious”
(internal quotation marks omitted)).
Appeal 2020-005211
Application 15/339,973

56
Moreover, contrary to Appellant’s argument, the claims do not merely
recite determining values such as distance and temporal proximity and
making simple comparisons between such values. Instead, the claims recite
generating certain types of “scores” based on particular recited information
(e.g., generating “a duplicate medication score” based on “the determination
of whether the medication is a valid duplicate or an invalid duplicate”);
generating an “aggregate score” based on the other scores; comparing the
aggregate score to various unnamed threshold scores; and determining
whether the aggregate medication listing is to be modified based on the
results. In short, the claims recite a desired result, i.e., a determination of
whether a modification is to be made to the medication listing based on
various factors, without the Specification having provided adequate
disclosure of how this result is to be achieved. Vasudevan, 782 F.3d at 682.
Appellant contends that the Specification “clearly describes a
cognitive computing system using reasoning algorithms to score various
evidence.” Appeal Br. 11 (citing Specification ¶¶ 62, 63, 72). Appellant
contends that the Specification “provides separate descriptions of the various
factors and features evaluated . . . and the generation of corresponding
scores,” as well as a description of “scores for various features[] weighted
relative to one another.” Id. at 11–13 (citing Specification ¶¶ 87–108, 159–
161). Appellant further contends that evaluation of an aggregate score and
comparison of a score or the aggregate score for a medication to one or more
threshold scores or a predetermined threshold score is also described. Id. at
13–14 (citing Specification ¶¶ 94–96, 110–112, 160–161, Fig. 5, original
claim 5).
Appeal 2020-005211
Application 15/339,973

57
We are not persuaded. Paragraphs 62, 63, and 72 provide a generic
description of how a Question Answering pipeline or system (QA system)—
“an artificial intelligence application executing on data processing hardware
that answers questions pertaining to a given subject-matter domain presented
in natural language”—evaluates candidate answers to “identify . . . answers
that surface as being significantly stronger than others and thus[] generates a
final answer, or ranked set of answers, for the input question.” Spec. ¶¶ 60–
63, 72. They do not provide any descriptions of the algorithm used to
achieve the particular desired results claimed (i.e., determining whether to
remove a medication from the medication list based on the enumerated
factors).
We acknowledge that the Specification provides some description of
how various recited factors (i.e., invalid duplicate medication,
contraindication, medication interaction, and cohort analysis) may contribute
to the desirability of removing a particular medication from a patient’s
medication list. Spec. ¶¶ 87–108, 159–161. Likewise, the Specification
provides some description of an “aggregating scoring embodiment . . . in
which scores are generated by each of the [duplicate medication analysis
engine, the contra-indicated medication engine, the medication interaction
analysis engine, and cohort analysis engine] and aggregated through a
weighted evaluation to generate an aggregate score that may be used to
determine whether to include or remove a medication from the patient’s
medication listing.” Spec. ¶ 110; see also Spec. ¶¶ 109–112, 159–161, Fig.
5, original claim 5; cf. Spec. ¶¶ 94–96 (describing weighting various factors
to determine whether a duplicate medication instance is valid or invalid).
Appeal 2020-005211
Application 15/339,973

58
However, as the Examiner points out, the Specification is inadequate
as to how these scores should be generated and weighted to determine
whether a medication should be removed from a patient’s medication listing.
Ans. 3. Instead, the Specification only states, for example, that “[t]he
weights applied to the various scores may be specified by a subject matter
expert, may be learned through a machine learning process, or the like.”
Spec. ¶ 160. We agree with the Examiner that such generic disclosures are
inadequate to describe the claimed method of determining whether or not the
aggregate medication listing data structure is to be modified.
Appellant contends that the “claimed invention is not limited to just
one type of rule/logic for determining . . . weighted scores.” Appeal Br. 12.
Appellant purports to provide example algorithms for generating recited
scores in the briefs. Appeal Br. 12. Appellant contends that,
just because a feature may be implemented in different
ways and the specification is written to cover any of the
various ways that one skilled in the art may implement
such features, this does not mean that the written
description is inadequate when these various ways of
implementing the particular calculations are within the
level of skill in the art given the teachings of the present
specification.
. . .
. . . While the specific numerical equations or
functions may be varied depending on the desired
implementation, this does not mean that one of skill in the
art would not have recognized Appellants as having
possession of the claimed invention at the time of filing.
Id. at 13–14; see also Reply Br. 2, 4 (“Just because there are multiple
different ways to weight values does not mean that Appellants did not have
possession of the claimed idea of weighing the values based on input data.”).
Appeal 2020-005211
Application 15/339,973

59
We are not persuaded. As discussed above, enablement and written
description are distinct requirements. Ariad, 598 F.3d at 1352–1353. “It is
not enough that one skilled in the art could theoretically write a program to
achieve the claimed function, rather the specification itself must explain how
the claimed function is achieved to demonstrate that the applicant had
possession of it.” 84 Fed. Reg. at 61–62 (citing Vasudevan, 782 F.3d at
682–83). Furthermore, the claims are not directed to the generic method of
weighing values based on input data. Instead, the claims are directed to a
method for determining whether to modify a patient’s medication listing
based on the recited factors. A skilled artisan reading Appellant’s generic
disclosures would not reasonably understand that the inventors had
possession of all possible implementations of achieving such a result.
Accordingly, we affirm the Examiner’s rejection of claims 27, 28, 35,
and 36 as invalid for lacking adequate description.
Claims 23, 24, 31, 32
Appellant contends that “persons have been determining the distance
between two points and comparing them to thresholds for centuries” and that
“[t]he claimed invention is not directed to how one determines a distance
between two points or how one compares such a distance to a threshold.”
Appeal Br. 15. Instead, Appellant contends, “the claimed invention is
directed to such a comparison of distance being part of the evaluation of
whether a medication instance is a valid duplicate medication or an invalid
duplicate medication.” Id.
We are not persuaded. We do not read the Examiner’s rejection as
being based on an assertion that a skilled artisan would not be able to
determine whether sources of a medication are within a predetermined
Appeal 2020-005211
Application 15/339,973

60
distance of a known location associated with a patient (claims 23, 31) or of
one another (claims 24, 32). Instead, the rejection is based on the finding
that, although the claims recite determining a medication is a valid duplicate
based on sources of a medication being within such predetermined distances,
there is not “a predetermined distance recited anywhere in the specification
that would adequately describe what a predetermined distance is indicative
of” in the context of such a determination. Ans. 5 (emphasis added).
Appellant points to paragraphs 87–89 and 95 of the Specification, as
well as original claims 6 and 16 as providing support for the limitation
regarding “predetermined distance” in claims 23, 24, 31, and 32. Reply Br.
4–5. These citations, however, support the Examiner’s rejection. Paragraph
87, for example, suggests information such as location of the source of
medication may be relevant to determining whether a medication is a
duplicate, with further analysis to be “performed to determine whether the
duplicate prescription is valid or not.” Spec. ¶¶ 87–88; see also id. ¶ 44.
Original claims 6 and 16 similarly suggest that “analysis of the . . .
geographical locations of the first and second sources” may be used to
“determin[e] that the medication conflict exists.” Spec. 76 (claim 6), 80
(claim 16). However, such disclosures do not indicate a relationship
between the location of the medication source and whether a medication is a
valid duplicate.
Likewise, paragraph 88 of the Specification teaches that,
[i]n order to verify the duplicate prescription, the location
of the source may be compared to other locations of
sources of medical services the patient has utilized, the
location of the patient’s home and work, and the like, to
determine if the location of a source of a duplicate
medication is relatively near these other locations. This
Appeal 2020-005211
Application 15/339,973

61
gives insight into whether the patient’s duplicate
prescription is potentially due to the patient having lost or
failed to take with them their original prescription, e.g., the
patient is on vacation or a trip away from their home
location and failed to bring with them their medication.
Spec. ¶ 88. Here, again, although the disclosure suggests the distance between
a medication source and “a known location associated with the patient” may
give “insight” into the reason for the potential duplication of medication, the
disclosure does not explain how such insight could or should be used to
determine whether a duplicate medication is valid or invalid.
Finally, paragraph 89 describes that similarity of location of the
source of two instances of the same prescribed medication, among other
things, may indicate that the latter instance of the medication is “a renewal
or additional prescription to supplement the previous prescription” from,
e.g., a patient’s primary care physician, and thus a valid duplicate. Spec.
¶ 89; see also id. ¶ 45. Accordingly, paragraph 89 at least provides some
description of the source locations of potential duplicate medications being
used to determine that a medication is a valid duplicate, as recited in claims
24 and 32.
Nevertheless, although we acknowledge that adequate written
description does not require disclosure of the claimed invention in haec
verba, the disclosure in paragraph 89 is inadequate to show that the
inventors had possession of the full scope of the invention of claims 24 and
32. In particular, although paragraph 89 suggests that “similarity” of source
locations may indicate that two instances of a medication are valid
duplicates, the paragraph does not suggest that any source located within a
predetermined distance would be considered “similar,” or otherwise teach
Appeal 2020-005211
Application 15/339,973

62
how location “within a predetermined distance” is related to “similarity” of
source location.
Finally, paragraph 95 also does not describe distance between a
medication source and “known location associated with the patient” being
used to determine that the medication is a valid duplicate (claims 23 and 31).
Instead, the Specification merely states that “[a] relatively lower weight
value may be given to an evaluation of the relative distance between the
location of the patient’s home and the source of medication distance, since
this is less likely to be indicative of a medication instance being duplicative
or not.” Spec. ¶ 95 (emphasis added).
In short, the Specification fails to provide details such that a skilled
artisan can reasonably conclude that the inventor had possession of the
invention of claims 23, 24, 31, and 32, i.e., where a medication is
determined to be a valid duplicate in response to the location of a medication
source being within a predetermined distance of a known location associated
with the patient or of a second source. Accordingly, we affirm the rejection
of claims 23, 24, 31, and 32 as invalid for inadequate written description.
Claims 22, 30
Appellant contends that
[a]t least paragraphs [0044], [0046], [0086]-[0088],
[0092]-[0093], [0154] describe the evaluation of the
proximity of dates and timing information of medication
instances. Paragraphs [0087]-[0094] as well as the other
paragraphs previously mentioned above describe the
weighting of features relative to one another to generate a
score indicative of whether or not one instance of a
medication present in the aggregate data is a duplicate of
another. These weighted scores are generated using rules
Appeal 2020-005211
Application 15/339,973

63
or logic, examples of which are given at least in
paragraphs [0088]-[0112] as discussed above.
Appeal Br. 16–17.
We acknowledge that the Specification teaches that timing
information of medications may be indicative of whether medication
instances are duplicates and whether a duplicate medication is valid. See,
e.g., Specification ¶¶ 44, 46, 86–89, 92–93, 154.
In particular, the Specification teaches that similar medication
prescribed from two different institution types . . . within
a short duration of each other (determined from dates
associated with the instances), are indicative of the
medication being duplicative, and potentially prescribed
for emergency or quick visits, e.g., one time use or limited
use while at a facility receiving treatment.
Spec. ¶ 88 (emphasis added). The Specification similarly teaches that,
in a case where there are two instances of the same
medication being prescribed, such as due to a follow up
visit with the patient’s primary care physician (PCP), a
location of the source of the instances will be similar, the
institution of the source will be similar, and the duration
of time between the instances of medication prescriptions
will be larger and may coincide with a previous
prescription expiration time (another feature that may be
extracted from the medication information in the patient
EMR data). This is indicative of the subsequent instance
of the medication prescription being a renewal or
additional prescription to supplement the previous
prescription and is not in fact a duplicate medication
prescription. Thus, while the medication instances may be
duplicative, they are valid duplicates.
Id. ¶ 89 (emphasis added).
Thus, the Specification describes two instances of “performing a . . .
determination . . . as to whether the medication is a valid or invalid duplicate
Appeal 2020-005211
Application 15/339,973

64
medication by evaluating a temporal proximity of the first and second
instance and differences in types and geographical location of the first
source and second source,” as recited in claims 22 and 30. Nevertheless, we
find that these two examples do not suffice to adequately describe the full
scope of the recited limitation in claims 22 and 30, which encompasses any
evaluation of temporal proximities of medication instances to determine
whether a medication is a valid or invalid duplicate.
Accordingly, we affirm the Examiner’s rejection of claims 22 and 30
as invalid for lack of adequate written description.

CONCLUSION
In summary:
Claim(s)
Rejected
35 U.S.C.
§
Reference(s)/Basis Affirmed Reversed
1, 10, 11,
20–36
101 Eligibility 1, 10, 11,
20–36

1, 10, 11, 20,
21, 25, 26,
29, 33, 34
103 Miller, Boone 1, 10, 11, 20,
21, 25, 26,
29, 33, 34

22, 23, 27,
28, 30, 31,
35, 36
103 Miller, Boone,
Biernacki
27, 28, 35,
36
22, 23,
30, 31
24, 32 103 Miller, Boone,
Biernacki, Arshad
24, 32
22–24, 27,
28, 30–32,
35, 36
112(a) Written
Description
22–24, 27,
28, 30–32,
35, 36

Overall
Outcome
1, 10, 11,
20–36

Appeal 2020-005211
Application 15/339,973

65
TIME PERIOD FOR RESPONSE
No time period for taking any subsequent action in connection with
this appeal may be extended under 37 C.F.R. § 1.136(a). See 37 C.F.R.
§ 1.136(a)(1)(iv).

AFFIRMED