Ex Parte BUELOW et al

Patent Trial and Appeal Board

Jan 25, 2019

14334176 (P.T.A.B. Jan. 25, 2019)

UNITED STA TES p A TENT AND TRADEMARK OFFICE
APPLICATION NO. FILING DATE FIRST NAMED INVENTOR
14/334,176 07/17/2014 THOMAS BUELOW
24737 7590 01/29/2019
PHILIPS INTELLECTUAL PROPERTY & STANDARDS
465 Columbus A venue
Suite 340
Valhalla, NY 10595
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
ATTORNEY DOCKET NO. CONFIRMATION NO.
2005P00340US 4812
EXAMINER
HOFFA, ANGELA MARIE
ART UNIT PAPER NUMBER
3793
NOTIFICATION DATE DELIVERY MODE
01/29/2019 ELECTRONIC
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.
Notice of the Office communication was sent electronically on above-indicated "Notification Date" to the
following e-mail address(es):
patti. demichele@Philips.com
marianne.fox@philips.com
katelyn.mulroy@philips.com
PTOL-90A (Rev. 04/07)
UNITED STATES PATENT AND TRADEMARK OFFICE
BEFORE THE PATENT TRIAL AND APPEAL BOARD
Ex parte THOMAS BUELOW, RAFAEL WIEMKER, CRISTIAN
LORENZ, STEFFEN RENISCH, and THOMAS BLAFFERT 1
Appeal2018-002744
Application 14/334, 17 6
Technology Center 3700
Before DANIEL S. SONG, JAMES P. CAL VE, and
AR THUR M. PESLAK, Administrative Patent Judges.
SONG, Administrative Patent Judge.
DECISION ON APPEAL
STATEMENT OF THE CASE
The Appellant appeals under 35 U.S.C. § 134(a) from the Examiner's
Final Office Action ("Final Act.") rejecting claims 1, 6-10, and 15-19 in the
present application. (App. Br. 10-12, Claims App.). We have jurisdiction
under 35 U.S.C. §§ 6(b) and 134(a).
We REVERSE.
1 The Appellant is the Applicant, Koninklijke Philips N.V., which is
identified as the real party in interest (Appeal Brief ("App. Br.") 2).
Appeal2018-002744
Application 14/334, 17 6
The claimed invention is directed to "[ a ]n imaging method for
identifying abnormal tissue in the lung." (Abstract). Representative
independent claim 1 reads as follows (App. Br. 10, Claims App., emphasis
added):
1. A method, comprising
recording slice images of a lung through X-ray radiation
to produce slice images of the lung, each slice image corresponds
to a different position of the lung, where at least one slice image
of the slice images includes lung cavity, blood vessels and an
abnormal tissue, wherein the lung cavity, the blood vessels and
the abnormal tissue are represented through gray-scale values;
segmenting the blood vessels in the slice images of the
lung, wherein the gray-scale values of the blood vessels are
differentiated from the gray-scale values of the abnormal tissue
through eigenvalues of a Hessian matrix;
removing the segmented blood vessels from the slice
images of the lung; and
displaying at least one of the slice images of the lung with
the removed segmented blood vessels on an output device.
Independent claim 10 is directed to a computer tomograph that
performs the method as substantially claimed in claim 1, and independent
claim 19 is directed to a computer readable storage medium having
instructions for causing a processor to perform the method as substantially
claimed in claim 1. (App. Br. 10, Claims App. 11-12).
2
Appeal2018-002744
Application 14/334, 17 6
REJECTION
The Examiner rejects claims 1, 6-10, and 15-192 under 35 U.S.C.
§ 103 as obvious over Lin et al. (US 2004/0151356 Al, pub. Aug. 5, 2004
("Lin")) in view of Reeves et al. (US 2004/0252870 Al, pub. Dec. 16, 2004
("Reeves")). (Final Act. 5).
ANALYSIS
In rejecting the claims, the Examiner finds that Lin discloses the
invention as substantially recited in claim 1, including wherein "the blood
vessels and the abnormal tissues are represented through gray-scale values"
and "identifying the blood vessels in the slice images of the lung ... wherein
the gray-scale values of the blood vessels are differentiated from the gray-
scale values of the abnormal tissue through eigenvalues of a Hessian
matrix." (Final Act. 5---6). In that regard, the Examiner finds the
identification of blood vessels to be satisfied by Lin's disclosure of
"enhancement filters; fig. 1 Ob shows enhancement of abnormal
tissue/nodules created in S23, fig. 2; fig. 10c shows enhancement of blood
vessels created in S25, fig. 2; par. 0133-0142 teach eigenvalues of a Hessian
matrix." (Id.). The Examiner also finds that
it is unclear if Li teaches "segmenting" the blood vessels. Image
segmentation is generally known in the art as partitioning image
pixels and classifying the pixels into defined segments. Li
demonstrates pixel assignment/segmentation of candidate
nodules (S28, fig. 2) but does not expressly teach segmentation
of entire blood vessels. Instead, only the blood vessel features
2 Claim 21 has been canceled. (App. Br. 10-12, Claims App.).
3
Appeal2018-002744
Application 14/334, 17 6
that overlap with the initial nodule candidates (S24, fig. 2) are
clearly segmented (S26, fig. 2).
(Final Act. 5-6; see also Answer ("Ans.") 3--4).
To remedy this deficiency, the Examiner finds that Reeves discloses
"segmentation of blood vessels from the C[ omputed] T[ omography] image
data (fig. 8, par. 0055, 0089)," and removing "segmented structures adjacent
to structures of interest (i.e. nodules)" to "enhance the display of the
structures of interest (i.e. nodules) (par. 0105)." (Final Act. 6). The
Examiner concludes that it would have been obvious to one of ordinary skill
in the art to have provided "the line-enhanced image of Li as a blood vessel
segmentation image in order to enhance the display of structures of interest
(i.e. nodules) as taught by Reeves." (Final Act. 6).
The Appellant argues that the rejection is inadequate because "Li
discloses segmenting nodules and not blood vessels." (App. Br. 5, citing Li
,r 142; see also App. Br. 6 ("Li cannot remove something that is not
detected"); Reply Br. 3). The Appellant also argues that "Li does not
disclose segmenting blood vessels and then removing the segmented blood
vessels from the image." (App. Br. 6, citing Li ,r 142).
However, these arguments are not persuasive in view of the
Appellant's Specification, which utilizes the term "segmentation" to refer to
the broader process of identifying particular structure from an image, for
removal, for example. (Spec. 1, 11. 27-29 ("known segmentation methods,
in which slice images of structures of the object to be examined are
subtracted from one another and individual structures are thereby removed
from the resulting image"); 5, 1. 33---6, 1. 1 ("After recording the tomograms,
interfering structures of the body are segmented, for instance, the heart or
4
Appeal2018-002744
Application 14/334, 17 6
the thoracic cage, that is, these are removed from the image"); 6, 1. 4--5
("After segmentation of the structures of the body, the lung cavity 5 and the
blood vessels 6 are left, as illustrated")).
The Examiner explains that "Li creates a blood vessel image (i.e. line
image) as shown in Fig. 1 OC. This can be considered a type of blood vessel
'segmentation' image because pixel values are assigned only to structures
representing lines (i.e. blood vessels)." (Ans. 6). This image is "compared to
the nodule candidate image" and "removed from the final nodule candidate
image." (Ans. 6). In order for the invention of Li to identify and display the
nodules, the remaining structures of the lung, including the blood vessels,
must be identified for removal. Accordingly, we agree with the Examiner that
"boundaries of blood vessels are implicitly determined in a segmentation
image to demonstrate blood vessel/not blood vessel interface." (Ans. 8).
The Appellant rebuts that "Li does not disclose, suggest or imply
segmenting blood vessels where the gray-scale values of the blood vessels
are differentiated from gray-scale values of abnormal tissue." (Reply Br. 2).
According to the Appellant, there is no indication that "line enhanced
image" or "image filtering" as disclosed in Li would be understood by those
of ordinary skill in the art as segmentation based on grey-scale values (Reply
Br. 2). We find the Appellant's contention to be persuasive and determine
that the rejection as presently set forth is inadequate.
While Li discloses segmentation in the sense of identifying non-nodule
structure in the lungs, it utilizes filtering to enhance images to do so such that
better identification of nodules can be attained. While the claims recite a
specific manner of attaining segmentation, Li discloses applying dot, line, and
5
Appeal2018-002744
Application 14/334, 17 6
plane filters to enhance the image and to distinguish nodules from other
structures within the lung. (Li ,r,r 44--47, 50-54, 140, 142). We do not
understand, nor has the Examiner adequately explained, how a filter-enhanced
image disclosed in Li is an image "wherein the gray-scale values of the blood
vessels are differentiated from the gray-scale values of the abnormal tissue"
as required by the claims.
The claims further require the displaying of "at least one slice image
of the lung with the removed segmented blood vessels on an output device."
(See e.g., App. Br. 10, Claims App. (claim 1)). It is unclear from the
Examiner's rejection where Li discloses such a display. The Examiner
explains that "Li suggests display of image of nodules (i.e. an image with
blood vessels removed) because it is the intended use of his invention. As
shown in Fig. 2, step S28 results in a nodule image wherein the blood
vessels have been removed." (Ans. 7). However, the image displayed are 3-
D images of just the nodules as generated using the disclosed enhancement
filters, and are not two-dimensional slice images of the lung with segmented
blood vessels removed as claimed.
The Appellant also argues that "Reeves does not cure the deficiencies
of Li." (App. Br. 5). We generally agree. While the Examiner explains that
Reeves "teaches that a blood vessel segmentation image, analogous to the
line image of Li, is subtracted from a lung image" such that "only nodule
candidates remain" (Ans. 6-7), such general disclosure, as applied by the
Examiner, does not adequately address the above-noted deficiencies of Li
pertaining to gray-scale, and display of at least one of the slice images of the
lung with the blood vessels removed. In that regard, the Appellant is correct
6
Appeal2018-002744
Application 14/334, 17 6
that Reeves eliminates "structures which are not nodules, which suggests
that lung cavity and nodules are not included together in any image contrary
to the claim limitations." (Reply Br. 4).
According to the Examiner:
[ A ]s evidenced by Li, Reeves, and the lack of detail in
Appellant's disclosure, one of ordinary skill would recognize
that image segmentation is highly routine in the art and provides
highly predictable results. The steps of the claimed method do
not appear to provide any technical advantage or benefit over that
which is taught by Li and Reeves and generally known by one of
ordinary skill in the image processing art.
(Ans. 7).
However, the Appellant is correct that a proper rejection is "not based
on advantage or benefit" (Reply Br. 3), but instead, must be based on the
claim language and the prior art as understood by those of ordinary skill.
Finally, the Examiner also responds that "[ o ]ne of ordinary skill in the
art would also find it routine to selectively highlight the desired tissues for
display (e.g. lung cavity, nodules, blood vessels)." (Ans. 8). While this
reasoning may be true, it is still inadequate to address the deficiencies of Li,
and the Examiner's rejection as noted above. To be clear, the record
indicates the use of gray-scale values to differentiate the lung from other
structures in the body appears to be well-known in the art. For example,
step S22 labeled "Lung Segmentation" in Figure 2 of Li discloses such a
technique. (See also Li ,r,r 140, 142). Nonetheless, the rejection, as
presently articulated, does not adequately address the actual claim language
pertaining to the recited segmentation and displaying.
7
Appeal2018-002744
Application 14/334, 17 6
As noted by the Appellant "[ c ]laims 10 and 19 include similar
limitations" to those of claim 1. (App. Br. 6). Therefore, in view of the
above considerations, we reverse the Examiner's rejections of claims 1, 10,
and 19. The remaining claims depend from one of these independent claims,
and as such, we also reverse the rejection of these claims as well.
CONCLUSION
The Examiner's rejection is REVERSED.
REVERSED
8