Ex Parte Tuma et al

Patent Trial and Appeal Board

Sep 17, 2012

11356737 (P.T.A.B. Sep. 17, 2012)

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.
11/356,737 02/17/2006 Gregor Tuma SCHWP0241USA 3671
7590 09/17/2012
RENNER, OTTO, BOISSELLE & SKLAR, LLP
Nineteenth Floor
1621 Euclid Avenue
Cleveland, OH 44115-2191
EXAMINER
DANEGA, RENEE A
ART UNIT PAPER NUMBER
3736
MAIL DATE DELIVERY MODE
09/17/2012 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 GREGOR TUMA, MARIO SCHUBERT,
FRANK GRUENSCHLAEGER, and MARTIN HAIMERL
__________

Appeal 2011-004810
Application 11/356,737
Technology Center 3700
__________

Before LORA M. GREEN, FRANCISCO C. PRATS, and
JACQUELINE WRIGHT BONILLA, Administrative Patent Judges.

PRATS, Administrative Patent Judge.

DECISION ON APPEAL
This appeal under 35 U.S.C. § 134 involves claims to a method for
determining a characteristic axis of a body structure. The Examiner entered
rejections for obviousness.
We have jurisdiction under 35 U.S.C. § 6(b). We affirm in part.
STATEMENT OF THE CASE
To ensure correct function of an artificial hip joint, “and to enable a
long service life for the artificial joint, the femoral component of the
artificial hip joint should be positioned in the bone structure of the femur in
a particular position with respect to the shaft axis and the femur neck axis”
Appeal 2011-004810
Application 11/356,737

2
(Spec. 1). Because two-dimensional x-ray images can be inadequate for
determining the three-dimensional coordinates required to correctly position
such implants, “computer tomography (CT) recordings of the body structure
in question are often generated. However, CT recordings subject the patient
to a significantly higher radiation exposure relative to the two-dimensional
x-ray recordings mentioned above” (id. at 2).
The Specification discloses “a method for determining characteristic
axes of a body structure, such as for determining the three-dimensional
position of the femur neck axis or the shaft axis of the upper leg bone or
femur” (id.). Preferably, “two or more two-dimensional recordings [are]
produced from different directions of each of a first area of the body
structure and optionally also a second, different area of the body structure.
The recordings, for example, can be two-dimensional x-ray recordings of the
areas of the body structure” (id.).
From these images, the “position of a characteristic axis, such as the
position of the shaft axis, for example, can be determined automatically,
manually or by means of an image processing algorithm” (id.). In one
embodiment, the “position of the shaft axis and, with the aid of a generic
model, the position of the femur neck axis can be determined, for example,
by rear-projection from preferably at least two recordings of the femur neck
and the shaft of a femur respectively” (id. at 4). Thus, “it is no longer
necessary to obtain a CT recording” (id.).
Claims 1-13 and 40 stand rejected and appealed (App. Br. 2). Claim 1
is representative reads as follows (emphasis added):

1. A method for determining a characteristic axis of a body
structure, comprising:
Appeal 2011-004810
Application 11/356,737

3
comparing each of at least two two-dimensional mappings of an
area of the body structure with a generic model of the area in question,
said generic model containing information on the position of the
characteristic axis;
ascertaining a mapping protocol for mapping the generic model
onto the respective mappings of the area;
using the ascertained mapping protocol to map the position of
the characteristic axis in the generic model onto each of the at least
two two-dimensional mappings of the area of the body structure; and
back-projecting the at least two two-dimensional mappings of
the area of the body structure to determine a three-dimensional
position of the characteristic axis.

The following rejections are before us for review:
(1) Claims 1, 3-7, 11-13, and 40, under 35 U.S.C. § 103(a) as obvious
over Bras1 (Ans. 3-5);
(2) Claim 2, under 35 U.S.C. § 103(a) as obvious over Bras and
Shimura2 (Ans. 5-6); and
(3) Claims 8-10, under 35 U.S.C. § 103(a) as obvious over Bras and
Krause3 (Ans. 7-9).
DISCUSSION
The Examiner found that Bras described a process substantially as
recited in claims 1 and 40, except that “Bras doesn’t expressly teach the axis
is positioned from the generic model onto the two images. However, Bras
teaches retro projecting areas of interest, such as contours onto the images as

1 A. Le Bras et al., Personalised 3D reconstruction of proximal femur from
low-dose digital biplanar radiographs, 1256 INTERNATIONAL CONGRESS
SERIES 214-219 (2003).
2 Shimura, U.S. Patent App. Pub. No. 2003/0176860 A1 (published
September 18, 2003).
3 Krause et al., U.S. Patent No. 6,711,432 B1 (issued March 23, 2004).
Appeal 2011-004810
Application 11/356,737

4
well as using the method to determine characteristic axes of the femur”
(Ans. 4). The Examiner reasoned, therefore, that an ordinary artisan would
have considered it obvious to “back-project those characteristics of interest
onto the two dimensional images to find the three dimensional characteristic
of interest based on this approach” (id.).
Appellants contend that “Bras does not teach back projecting two-
dimensional mappings of the area of the body” as required by claim 1 (App.
Br. 6). Rather, Appellants urge, given Bras’ disclosure that its “‘algorithm
performs an elastic deformation of the generic object to fit its 3D retro-
projected contours with the contours identified on both radiographs,’. . .
[c]learly, retro-projection as disclosed in Bras applies to the ‘generic object’,
and not the radiographs (i.e., the alleged two-dimensional mappings)” (id. at
7-8 (quoting Bras 216)). Appellants reason, therefore, that “Bras does not
teach or suggest back-projecting the at least two two-dimensional mappings
of the area of the body structure to determine a three-dimensional position
of the characteristic axis as set forth in claim 1” (id.; see also Reply Br. 2-3).
As stated in In re Oetiker, 977 F.2d 1443, 1445 (Fed. Cir. 1992):
[T]he examiner bears the initial burden . . . of presenting a
prima facie case of unpatentability. . . .
After evidence or argument is submitted by the applicant
in response, patentability is determined on the totality of the
record, by a preponderance of evidence with due consideration
to persuasiveness of argument.

Appellants’ arguments do not persuade us that a preponderance of the
evidence fails to support the Examiner’s prima facie case as to claim 1.
Bras describes a process of producing a three-dimensional
reconstruction of a proximal femur from low dose biplanar radiographs (see
Appeal 2011-004810
Application 11/356,737

5
Bras 214 (abstract); see also id. at 215 (“The low dose digital X-ray device
EOTM performs a linear scanning of each femur in two different incidences
(posterior-anterior 0° and lateral views).”). As required by claim 1, using an
ascertained mapping protocol, “the NSCC algorithm,” Bras compared a “3D
generic model” to anatomic contour features identified in the radiographs
(see id. at 216).
Bras also discloses using the mapping protocol to fit the structures
identified in the 3D generic model onto the contours identified in the
radiographs (id. (“[T]he NSCC algorithm performs an elastic deformation of
the generic object to fit its 3D retro-projected contours with contours
identified on both radiographs.”)). Given Bras’ disclosure that its methods
are useful for calculating hip axis length, which would be useful “for
planning of surgical acts or computed assisted surgery” (Bras 214 (abstract);
see also id. at 219), we discern no error in the Examiner’s finding that a
surgeon of ordinary skill would have been prompted to apply Bras’
methodology to map the location of a characteristic axis in the generic
model onto the two dimensional radiographs, as also required by claim 1.
As to the back-projecting step at issue, we note Bras’ disclosure that
“[a]n iterative control of the reconstructed models is performed by
visualizing the retro-projection of the 3D reconstructed model on the real
femoral contours seen on both radiographs” (id. at 216-17 (emphasis
added)). As Bras explains, “[i]f the retro-projected contours are close to the
real radiograph contours, the 3D model is accepted. If not, contours are
modified in order to match the retro-projected model contours with the
radiograph contours” (id. at 217).
Appeal 2011-004810
Application 11/356,737

6
Thus, while it may be true that Bras’ process involves a retro-
projection of the 3D generic object, Bras’ retro-projection itself includes or
at least suggests the use of both of the two radiographs, as required by claim
1. Appellants do not point to any specific language in claim 1, or to any
clear or specific supporting disclosure which defines or explains the back-
projecting step, that would exclude a combined retro-projection of the 3D
generic model and the two-dimensional radiographs to determine the three-
dimensional position of a characteristic axis, as suggested by Bras.
Appellants’ arguments therefore do not persuade us that the Examiner erred
in finding that Bras suggests the claimed back-projecting step.
Thus, as Appellants’ arguments do not persuade us that the Examiner
failed to make out a prima facie case of obviousness as to claim 1, we affirm
the Examiner’s rejection of that claim over Bras. As they were not argued
separately, claims 3-6, 11, and 12 fall with claim 1. See 37 C.F.R.
§ 41.37(c)(1)(vii).
Appellants’ arguments also fail to persuade us that the preponderance
of the evidence does not support the Examiner’s prima facie case as to claim
7. Claim 7 recites “the method according to claim [1, further comprising
extracting contours of the body structure,] wherein extracting includes using
the method of orthogonal curves” (App. Br. 16).
Appellants argue that Bras “is silent with respect to using orthogonal
curves to extract the contours” (id. at 9). The Examiner finds, however, that
“since curves are extracted using two images taken at perpendicular angles
to each other, to create a three dimensional model, the model [in Bras] will
be based on using orthogonal curves or curves from each perpendicular
image” (Ans. 8).
Appeal 2011-004810
Application 11/356,737

7
Appellants do not dispute the Examiner’s finding. Moreover,
Appellants’ Specification discloses that the use of orthogonal curves was a
known edge detection method (see Spec. 3-4). Appellants’ arguments do
not, therefore, persuade us that a preponderance of the evidence fails to
support the Examiner’s conclusion of obviousness as to claim 7. We
therefore also affirm the Examiner’s obviousness rejection of claim 7 over
Bras.
As to claim 13, which depends from claim 1, Appellants reiterate their
argument, discussed above, that Bras retro-projects the contours of the
generic object, rather than the two radiographs, and urge, therefore, that “[i]f
Bras does not perform retro-projection on the alleged two-dimensional
mappings (i.e., the radiographs), it follows that Bras also does not produce a
three-dimensional body structure by back-projecting the two-dimensional
mappings” (App. Br. 10). Furthermore, Appellants argue, “the cited text of
Bras is silent with respect to calculating boundary lines or boundary areas of
such three-dimensional body structure model from back-projected body
contour mappings” (id.).
We are not persuaded. As discussed above, Appellants’ arguments do
not convince us that claim 1 fails to encompass the retro-projecting steps
described in Bras.
Moreover, given Bras’ production of a 3D model in which “contours
are generated as if they were projected on postero-anterior and lateral
radiographs” (Bras 216 (Fig. 1 caption)), we discern no error in the
Examiner’s finding that Bras’ method includes a step of generating a three-
dimensional model as required by claim 13. Also, given the use of the
NSCC algorithm to retrofit the contours from the generic 3D model to the
Appeal 2011-004810
Application 11/356,737

8
radiographs, we are not persuaded that Bras’ method fails to include claim
13’s calculation of boundary lines from back-projected contour mappings.
We therefore also affirm the Examiner’s rejection of claim 13 as obvious
over Bras.
We agree with Appellants, however, that the Examiner has not
adequately explained why an ordinary artisan would have considered claim
40 to be obvious over Bras.
Claim 40 recites, “[t]he method according to claim 1, further
comprising using an intersecting line of back-projection planes ascertained
from back-projecting the at least two two-dimensional mappings of the area
to define the three-dimensional position of the characteristic axis” (App. Br.
16-17). As the Examiner points to no clear or specific teaching in Bras
suggesting that a characteristic axis should be defined using an intersecting
line of the back-projection planes of the two two-dimensional mappings, we
reverse the Examiner’s rejection of claim 40 over Bras.
The Examiner rejected claim 2 over Bras and Shimura, reasoning that
an ordinary artisan would have considered it obvious to use Bras’ method to
calculate the position of a second characteristic axis, in view of Shimura’s
teaching of determining the location of two characteristic axes when
performing prosthesis deployment surgery (see Ans. 5-6). Like claim 40,
however, claim 2 requires the second axis’ position to be determined using a
“calculation comprising an intersecting straight line of the back-projection
planes of the central axes of the body contours in the two-dimensional
mappings” (App. Br. 15). Because the Examiner points to no clear or
specific teaching in either Bras or Shimura suggesting that a characteristic
axis should be defined using an intersecting line of the back-projection
Appeal 2011-004810
Application 11/356,737

9
planes of the two two-dimensional mappings, we reverse the Examiner’s
rejection of claim 2 over those references.
The Examiner also rejected claims 8-10, which all depend directly or
ultimately from claim 1, as obvious over Bras and Krause (Ans. 7-9). The
Examiner conceded that Bras did not expressly teach pre-registering
landmarks on the body before generating the mappings, as required by claim
8 (id. at 6).
To address that deficiency, the Examiner cited Krause as teaching “a
method of identifying characteristic axes during robotic surgery in which the
body part had preregistered landmarks to assist a surgeon or robot in
comparing the actual body with the model during surgery” (id. (citing
Krause, col. 5, ll. 15-25)). Based on these teachings, the Examiner
concluded that an ordinary artisan would have considered it obvious “to pre-
register landmarks on the body structure in Bras in order to use the method
in robotic guided surgeries” (id.).
Appellants argue that Krause discloses registering the markers “to a
pre-surgical 3D model. A pre-surgical model infers that image data (e.g.,
any two-dimensional recording) used to construct such model has already
been obtained” (App. Br. 14). Thus, Appellants urge, it cannot reasonably
be concluded that “combining the teachings of Bras and Krause results in
pre-registering the body structure before generating the at least two two-
dimesnional [sic] mappings. Instead, and assuming Bras and Krause are
combined as suggested by the Examiner, it is logical to conclude such
registration occurs after the mappings are obtained” (id.).
The Examiner responds that “Krause specifically contemplates images
taken of the patient after marker registration or placement to calculate, or
Appeal 2011-004810
Application 11/356,737

10
adjust, the final surgical plan” (Ans. 9 (citing Krause, abstract)). Thus, the
Examiner urges, since “Bras’s 3-D model exists even before the x-ray
images are taken, the methods of Krause can be used to further accurately
define landmarks on the model, the x-ray images, and the body for use
during surgery” (id.).
We find that the Examiner has the better position.
Claim 8 recites “[t]he method according to claim 1, further
comprising pre-registering the body structure before generating the
mappings” (App. Br. 16). Thus claim 8 encompasses placement of
anatomical registration markers before the mapping protocol recited in claim
1 is used to map the position of the characteristic axis in the generic model
onto the two two-dimensional mappings.
We acknowledge that Krause describes using “an IR marker system
. . . during surgery . . . [which] allows the surgeon to more accurately follow
the surgical plan” (Krause, col. 5, ll. 16-25). As the Examiner points out,
however, Krause more broadly discloses a computer aided orthopedic
surgery method that involves using software to generate a 3D model of the
relevant area of the body (id. at abstract).
Specifically, Krause discloses that the “software model is used to
generate a surgical plan, including placement of multifunctional markers, for
performing the surgical procedure. After the markers are placed on the
patient, an updated image of the patient is taken and used to calculate a final
surgical plan for performing the remainder of the surgical procedure” (id.).
Given Krause’s disclosure of placing registration markers before
taking updated images of the patient, and before calculating the final
surgical plan, we are not persuaded that the Examiner erred in finding that
Appeal 2011-004810
Application 11/356,737

11
Krause would have suggested to an ordinary artisan practicing Bras’
methods that it would be useful to pre-register anatomical features in the
body structure of interest before mapping it, so as to assist in the preparation
of Bras’ ultimate mapping protocol. Thus, as Appellants’ arguments do not
persuade us that the preponderance of the evidence fails to support the
Examiner’s prima facie case as to claim 8, we affirm the Examiner’s
rejection of that claim over Bars and Krause. Claims 9 and 10 fall with
claim 8 as they were not argued separately. See 37 C.F.R. § 41.37(c)(1)(vii).

SUMMARY
We affirm the Examiner’s rejection of claims 1, 3-7, and 11-13 as
obvious over Bras.
However, we reverse the Examiner’s rejection of claim 40 as obvious
over Bras.
We also reverse the Examiner’s rejection of claim 2 as obvious over
Bras and Shimura.
However, we affirm the Examiner’s rejection of claims 8-10 as
obvious over Bras and Krause.

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

AFFIRMED-IN-PART
lp