Align Technology, Inc.

Patent Trials and Appeals Board

Jul 6, 2020

IPR2019-00148 (P.T.A.B. Jul. 6, 2020)

Trials@uspto.gov Paper 28
571-272-7822 Date: July 6, 2020

UNITED STATES PATENT AND TRADEMARK OFFICE

BEFORE THE PATENT TRIAL AND APPEAL BOARD

3SHAPE A/S and 3SHAPE INC.,
Petitioner,
v.
ALIGN TECHNOLOGY, INC.,
Patent Owner.

IPR2019-00148
Patent 9,451,873 B1

Before BRIAN J. McNAMARA, NEIL T. POWELL, and
ELIZABETH M. ROESEL, Administrative Patent Judges.
McNAMARA, Administrative Patent Judge.

JUDGMENT
Final Written Decision
Determining All Challenged Claims Unpatentable
35 U.S.C. § 318(a)

IPR2019-00148
Patent 9,451,873 B1
2
BACKGROUND
On July 12, 2019, we instituted an inter partes review of claims 1–20
of U. S. Patent No. 9,451,873 B1 (“the ’873 Patent”). Paper 9 (“Dec. to
Inst.”). Patent Owner filed a Patent Owner Response (Paper 12, “PO
Resp.”), Petitioner filed a Petitioner Reply (Paper 17, “Pet. Reply”), Patent
Owner filed a Patent Owner Surreply (Paper 18, “PO Surreply”) and a
transcript of an oral hearing held on April 15, 2020 (Paper 27, “Hr’g. Tr.”)
has been entered into the record.
We have jurisdiction under 35 U.S.C. § 6. This Final Written
Decision is issued pursuant to 35 U.S.C. §318(a). We base our decision on
the preponderance of the evidence. 35 U.S.C. § 316(e); 37 C.F.R. § 42.1(d).
Having reviewed the arguments of the parties and the supporting
evidence, we conclude that Petitioner has demonstrated by a preponderance
of the evidence that the challenged claims are unpatentable.
THE ’873 PATENT (EX. 1001)
The ’873 patent concerns prosthodontics procedures using intraoral
scans of intraoral sites and 3D models to implant a dental prosthesis, such as
a crown, denture or bridge, in an oral cavity. Ex. 1001, 1:13–18, 2:13–15,
7:31–35. In many prosthodontics procedures, an existing tooth, known as
the preparation tooth, is ground down to a stump so the prosthesis can be
mounted or seated on the preparation tooth. Id. at 5:17–26. The border
between a natural (unground) portion of the preparation tooth and the
prepared (ground) portion of the preparation tooth is called the finish line.
Id. at 5:20–24. In many instances the finish line is below the gum line. Id.
at 5:28– 29. Where necessary to expose the finish line, a practitioner inserts
a cord between the preparation tooth and the patient’s gum, then removes the
cord and scans and intraoral image before the soft tissue of the gum reverts
IPR2019-00148
Patent 9,451,873 B1
3
back to its natural position. Id. at 5:43–54. When satisfied with the intraoral
image of the preparation tooth, the practitioner obtains intraoral images of
adjacent areas to ensure the dental prosthesis will fit in the patient’s mouth.
Id. at 5:55–67. The additional sets of intraoral images may capture portions
of the preparation tooth, as well. Id. at 5:66– 6:3.
In one embodiment of the invention, an “image locking module”
automatically locks a first set of intraoral images associated with a
preparation tooth, without locking other image data sets. Id. at 6:4–15. The
image locking module may use the identity of the preparation tooth to select
automatically portions of the locked first set of intraoral images that will be
applied by the 3D model. Id. at 6:16–19. Alternatively, a practitioner can
use a graphical user interface (GUI) to mark portions of the locked set to be
applied to the 3D model. Id. at 6:19–22. Using the locked data
representing the preparation tooth in the 3D model, while ignoring lower
quality depictions of the preparation tooth, avoids degrading the finish line
captured in the first image data. Id. at 6:43–48. According to the ’873
patent, all the data within the finish line plus a 1–3 mm buffer around the
finish line may be locked for use in the 3D model. Id. at 6:30–35. The
determination of what data to keep in the locked image data set may be
accomplished algorithmically or by the user through a GUI. Id. at 6:35–42.
The image locking module associates each intraoral image data set
with a particular tooth “and/or may otherwise identify an intraoral site
associated with each intraoral image data set.” Ex. 1001, 7:16–20. A user
may indicate a tooth he or she is scanning before generating an image set or
take the image data set and subsequently indicate the tooth imaged. Id. at
7:20–24. A scan session is complete when all images for an intraoral site
have been captured. Id. at 7:31–32. To generate a virtual 3D model, a
IPR2019-00148
Patent 9,451,873 B1
4
“model generation module” “may register (i.e., ‘stitch’ together) the
intraoral images generated from the intraoral scan sessions” by capturing 3D
data of various surface points in multiple images and computing
transformations between the images, e.g., transformations which align one
image with another by fitting the points of the each image and using local
searches around points to match points of two adjacent images. Id. at 7:32–
54.
Figure 2 of the ’837 patent is reproduced below.

Figure 2 of the ’873 patent
Figure 2 illustrates a flow diagram for method 200 of automatically locking
an image set of an intraoral site. Ex. 1001, 13:27–29. Figure 2 shows that
upon starting a scan session at step 205, a dental practitioner uses an
intraoral scanner to create a set of images focused on a particular intraoral
IPR2019-00148
Patent 9,451,873 B1
5
site, e.g., focused on a particular tooth. Id. at 13:29–34. Processing logic
may direct the dental practitioner to a particular site, or the practitioner may
indicate the intraoral site to be scanned; alternatively, processing logic may
automatically (e.g., algorithmically) identify the intraoral site based on the
data from the set of images or other criteria. Id. at 13:31–42. At step 210
processing logic receives a set of intraoral images of the intraoral site. Id. at
13:42–43. Step 212 shown in Figure 2 as “determine the identity of the
intraoral site” is not discussed explicitly in the Specification of the ’873
patent. At step 215, the processing logic “locks” the intraoral image data to
ensure that portions of the intraoral image data set depicted, e.g., a particular
preparation tooth, will not later be degraded by additional images. Id. at
13:43–48.
Figure 7A of the ’837 patent shown below further illustrates the
process of Figure 2.

Figure 7A of the ’873 patent
Figure 7A shows a portion of dental arch 700 in which preparation teeth 708
and 712 have been ground to stumps as abutments for a dental bridge.
Ex. 1001, 13:49–54. In the intraoral image set 713 of preparation tooth 708,
intraoral images 714, 716, and 718 are generated by scanning at a particular
IPR2019-00148
Patent 9,451,873 B1
6
position (scanner station), preferrably to provide an overlap among images
714–718. Id. at 13:63–14:11. Intraoral image data set 713 is locked
automatically. Id. at 14:25–27.
At block 220 of Figure 2 discussed above, selected portions identified
using image analysis or other techniques correspond to a contour or other
feature of the intraoral site. Id. at 14:28–40. For example, as shown in
Figure 7A, intraoral images 714–718 of image set 713 inside a contour are
secured or “locked” to prevent further alteration, although the area outside
the contour remains unlocked. Id. at 14:41–48. In one embodiment,
reference data may be used to train a processing device to detect particular
objects, e.g., teeth, such that the known identity of a tooth or intraoral site is
used to assist in object detection and to select portions of the intraoral
images. Id. at 14:36–40.
At step 225 the processing logic receives additional intraoral images
of the intraoral site, e.g., images that may be part of one or more image sets
of additional teeth. Id. at 14:49–55. At step 230, the processing logic
generates a virtual 3D model that includes the intraoral site. According to
the ’873 patent:
The selected portions of the locked intraoral image (e.g., that
are inside of the determined contour) are used to create a first
region of the model. For example, the selected portions may be
used to create a particular preparation tooth in the 3D model.
Data from the additional intraoral images are not used to create
the region of the 3D model.
Id. at 14:56–62.
Figure 7B of the ’873 patent is shown below.
IPR2019-00148
Patent 9,451,873 B1
7

Figure 7B of the ’837 patent
Figure 7B illustrates the dental arch of Figure 7A after a second intraoral
image set 721 for tooth 706 adjacent to preparation tooth 708. Ex. 1001,
14:63–66. In Figure 7B, intraoral image set 721 includes intraoral images
722, 724, 726, and 728 that focus on adjacent tooth 706, although an area of
image 726 includes a portion of preparation tooth 708 and its finish line 709.
Id. at 14:66–15:3. Notwithstanding the overlap, data from intraoral image
set 721 will not be used in creating a virtual image of preparation tooth 708
because intraoral image set 713 for tooth 708 has been locked. Id. at 15:4–7.
ILLUSTRATIVE CLAIM
Claims 1, 11 and 20 of the ’873 patent are independent. Illustrative
claim 1 is drawn to a method and reproduced below with the claim limitation
reference designations used by Petitioner:
1. A method, comprising:
[1.1] receiving an intraoral image of a first intraoral site;
[1.2] determining an identity of the first intraoral site;
[1.3] algorithmically performing the following by a processing
device:
[1.4] locking the intraoral image; and
IPR2019-00148
Patent 9,451,873 B1
8
[1.5] selecting, based at least in part on the identity of the first
intraoral site, a portion of the intraoral image depicting a
portion of the first intraoral site; and
[1.6] generating a model comprising the first intraoral site
based at least in part on the locked intraoral image,
wherein the portion of the locked intraoral image is used
for a first region of the model, and wherein data from one
or more additional intraoral images that also depict the
portion of the first intraoral site is not used for the first
region of the model.
Independent claim 11 is drawn to a non-transitory storage medium
having instructions that cause a processing device to perform the operations
recited in the claim limitations. The operations recited in claim 11 are
similar to the method steps recited in claim 1, but include receiving a first set
of oral images of a preparation tooth showing a finish line of the preparation
tooth and corresponding limitations on the locking, selecting, and generating
operations. Independent claim 20 is drawn to a computing device
comprising a memory and a computing device to perform operations similar
to those recited in claim 1. Claims 11 and 20 explicitly recite that the
locking and selecting operations are performed without user input.

GROUNDS OF INSTITUTION
The sole challenge in the Petition is that claims 1–20 are unpatentable
as obvious under 35 U.S.C. § 103(a) over Kopelman1 in view of Paley.2 Pet.
4.

1 PCT Publication No. WO 2012/011101 (Ex. 1003)(“Kopelman”))
2 U.S. Patent Publication No. 2007/0172112 (Ex. 1004)(“Paley”)
IPR2019-00148
Patent 9,451,873 B1
9
CLAIM CONSTRUCTION
The Petition has been accorded a filing date of November 10, 2018.
Paper 6. For Petitions filed before November 13, 2018, we interpret claims
of an unexpired patent using the broadest reasonable construction in light of
the specification of the patent in which they appear. See 37 C.F.R.
§ 42.100(b)(2018); Cuozzo Speed Techs. LLC v. Lee, 136 S. Ct. 2131, 2144–
46 (2016). In applying a broadest reasonable construction, claim terms are
generally given their ordinary and customary meaning, as would be
understood by one of ordinary skill in the art in the context of the entire
disclosure. See In re Translogic Tech., Inc., 504 F.3d 1249, 1257
(Fed. Cir. 2007). Any special definition for a claim term must be set forth in
the specification with reasonable clarity, deliberateness, and precision. In re
Paulsen, 30 F.3d 1475, 1480 (Fed. Cir. 1994).
In the Preliminary Response, Patent Owner asserted that no claim
terms required construction and that all terms should be given their plain and
ordinary meaning. Dec. to Inst. 7 (citing Prelim. Resp. 18). As proposed by
Petitioner we construed “intraoral images” to mean “scanned images
obtained directly from an intraoral cavity of a patient,” “algorithmically
performing” to mean “automatically, without user input,” and “locking” to
mean “storing image data and disregarding any data from other images that
spatially overlap the ‘locked’ image.” Id. at 8–9. Except for the term
“locking,” neither party disputes the claim constructions we applied in the
Decision on Institution.
The Patent Owner Response proposes constructions of additional
terms and a different construction of the term “locking.” We address Patent
Owner’s newly proposed constructions below.
IPR2019-00148
Patent 9,451,873 B1
10
1. Identity of an intraoral site (claim 1 and 20); identity of a
preparation tooth (claim 11)
Patent Owner proposes that we construe “identity of an intraoral site”
(in claims 1 and 29) and “identity of a preparation tooth” (in claim 11) (“the
identity term”) to mean “designation of a tooth or portion of a jaw that
specifies a location in the intraoral space.” PO Resp. 17.
Petitioner contends that Patent Owner is attempting to narrow the
scope of the claims, that there is no need to construe this term, and that we
should apply the plain and ordinary meaning. Pet. Reply 4–8. Petitioner
does not define the plain and ordinary meaning.
Patent Owner argues that both parties describe the claimed method of
curating scan image data using the identities of intraoral sites as reference
points for locking certain scan images to particular locations in the model of
the intraoral space. PO Resp. 18. According to Patent Owner, consistent
with this description, Petitioner’s expert, Dr. Saber, agreed the plain and
ordinary meanings of “an identity of an intraoral site” includes a designation
of a location of an intraoral site in an intraoral cavity, i.e., a tooth number in
a dental numbering system. Id. (citing Ex. 2007, Saber Tr. 47:15–19).
When asked if the identity of a preparation tooth could be the “number of
the tooth in a numbering convention,” Dr. Saber actually testified “that’s
possible. It depends on the context of the application.” Ex. 2007, 47:15–19.
Thus, we find that Dr. Saber did not concede that a numbering convention is
the only method of specifying the identity of an intraoral site, as that term is
used in the claims.
According to Patent Owner, the ’873 patent requires determining the
identity of the intraoral site before generating the virtual model, because the
model is generated in part based on the identity determinations. PO Resp.
IPR2019-00148
Patent 9,451,873 B1
11
19. As discussed further herein, Patent Owner argues this is important
because Kopelman teaches performing a suitability analysis after the model
is generated. Id. Patent Owner criticizes Dr. Saber’s assertion that analysis
to determine the suitability of the first physical part for the purpose of
allowing a prosthesis to be seated determines the identity of the first oral site
(i.e., the tooth abutment that will receive the prosthesis). Id. (citing Ex. 1024
¶ 42)3. Patent Owner contends Dr. Saber’s analysis incorrectly subsumes
the concept of identity within suitability. Id. Patent Owner argues
determining suitability does not involve using the intraoral site identity as a
reference point in the computer alogorithm to which image data can be
locked. Id.
We agree with Patent Owner that one cannot perform the steps of
“selecting, based at least in part on the identity of the first intraoral site, a
portion of the intraoral site depicting a portion of the first intraoral site”
without first determining the identity of the site. However we are not
persuaded by Patent Owner’s characterization of Dr. Saber’s testimony
because that testimony goes to the reason why or how a particular intraoral
site is selected by the user or computer for further analysis, not to
determining the identity of that intraoral site.
The specification describes an intraoral site as, e.g., a “dental site”
(Ex. 1001, 2:15), a “dental site within the oral cavity” (id. at 5:1–2), “a
portion of a jaw” (id. 2:57), a “tooth” or “an intraoral site associated with
each intraoral image data set” (id. at 7:17–20) and a site identified “based on
data from the set of intraoral images.” It appears that both parties agree that

3 Patent Owner’s citation to Ex. 1024 ¶ 42 is incorrect. The testimony
quoted by Patent Owner appears at Ex. 1024 ¶ 85.
IPR2019-00148
Patent 9,451,873 B1
12
some form of identification of the intraoral site is required. The claims
recite “determining the identity of the introral site.” The claims do not
specify any particular form of identity and do not require that the intraoral
site be suitable for any particular purpose. For example, Patent Owner
agrees that one way of determining the identity of a tooth is to examine its
shape. Hr’g Tr. 21–22.
As noted above, the claims also do not specify a particular order of
steps, except to the extent that the step reciting “selecting, based in part of
the identity of the first introral site, a portion of the intraoral image depicting
a portion of the intraoral site’ (claim limitation 1.5) implies that the identity
of the intraoral site was determined previously. As to the locking step,
Petitioner emphasizes the permissive language of the ’873 patent. Pet. 6–9.
See Ex. 1001, 6:16–19 (“The identity of the preparation tooth may be used
by image locking module 128 to automatically select portions of the locked
first set of intraoral images that will be applied for the preparation tooth in a
3D model. Alternatively, a practitioner may use a graphical user interface
(GUI) to mark the portions of the locked set of intraoral images that will be
applied for the preparation tooth in the 3D model.”) Petitioner further
argues that Patent Owner seeks to unduly complicate the step of determining
the identity of the first intraoral site. Pet. Reply 7–8. We find that, although
the “suitability” of a site for any particular purpose may be the reason a site
is selected for analysis or scanning, the suitability of the intraoral site for a
particular purpose is not its identity.
As noted above, Step 212, shown in Figure 2 of the ’873 patent as
“determine the identity of the intraoral site,” is not discussed explicitly in the
’873 patent Specification. Patent Owner contends that the “identity” of an
intraoral site is properly construed as “a designation of a tooth or portion of
IPR2019-00148
Patent 9,451,873 B1
13
a jaw that specifies a location in the intraoral space.” PO Resp. 17.
Petitioner agrees that there is no dispute as to what identity means, “[a]s
long as that construction would not exclude some kind of physical marking
in the image or model.” Hr’g. Tr. 8.
The ’873 patent states an intraoral image set may be associated with a
particular tooth or otherwise be associated with an intraoral image data set,
and that processing logic may identify the intraoral site based on data from a
set of intraoral images and/or one or more additional sets of images. Ex.
1001, 7:17–20, 13:37–42. In the context of the ’873 patent, the evidence
does not support a conclusion that the identity of an intraoral site must be
specified in any particular way. Therefore, we find that any form or format
of identification that uniquely designates a tooth or portion of a jaw in the
intraoral space is satisfactory. Thus, we find that “identity of the first
intraoral site,” as recited in claim 1, means any representation that uniquely
specifies a location in the intraoral cavity, including, but not limited to a
numbering system, a mark, an intraoral image data set, or a designation
based on a description of the site and its surroundings, e.g. a description of a
tooth having a particular shape.
2. Locking
In our Decision to Institute, consistent with Petitoiner’s proposed
construction, we construed “locking” to mean “storing image data and
disregarding any data from other images that spatially overlap the ‘locked’
image data.” Dec. to Inst. 8.
In its Preliminary Response, Patent Owner argued no construction of
“locking” is required. See Dec. to Inst. 8 (citing Prelim. Resp. 18). Patent
Owner now contends we should construe “locking” an image or set of
images to mean “exclusively associating data from an image or set of images
IPR2019-00148
Patent 9,451,873 B1
14
with an identity of an intraoral site.” PO Resp. 20. According to Patent
Owner, the ’837 patent Specification requires the locking to be
(1) associated with a particular location, e.g., a particular tooth, and
(2) associated exclusively with the identity of the site, “meaning [that] the
algorithm will prioritize images data locked to a particular site to the
exclusion of overlapping image data not locked to that site when rendering
that site in the model.” PO Resp. 21 (citing Ex. 1001, 8:48–53).
Petitioner contends that Patent Owner seeks to unduly limit the
construction of locking based on optional aspects of the Specification and to
insert a location requirement. Pet. Reply 9–10. According to Petitioner,
Patent Owner’s proposed construction lacks support in the Specification and
“[l]ocking the intraoral image in the ’873 patent simply means that data from
other images that overlap the ‘locked image’ (i.e., stored image data) is
discarded.” Id. at 10 (citing Ex. 1001, 2”40–49). Petitioner argues that the
873 patent does not tie “locking” to the identity of the intraoral site. Id.
Patent Owner notes that the ’873 patent states intraoral images may be
associated with a particular tooth or a preparation tooth and that image
locking module 128 may use the identity of an associated tooth to determine
what portions of that image set will be used exclusively for creation of a
particular region of a 3D model. PO Surreply 21–22 (citing Ex. 1001,7:24–
30, 8:48–53). Patent Owner cites the testimony of Petitioner’s expert that if
an image was locked to tooth 708, that image would not be used to render
tooth 710, as evidence supporting Patent Owner’s view that the term
“locking” is used and understood to refer to the process of associating
certain images and portions of images with particular intraoral sites. PO
Surreply 24 (citing Ex. 2007, Saber Dep. Tr., 57:9–21). We find that Patent
Owner’s assertion neglects the premise of the question posed to Dr. Saber,
IPR2019-00148
Patent 9,451,873 B1
15
i.e., that tooth 708 had been scanned and the scanned image had been locked
to that tooth. When viewed in context, we find that Dr. Saber’s testimony
does not support Patent Owner’s argument that there exists a specific
relationship between particular intraoral sites and images or portions of
images. The testimony stands only for the unremarkable principle that the
image data obtained from scanning a particular tooth is the image data for
that tooth. See also, Ex. 1001, 7:17–30 (describing various options for
identifying intraoral sites and associating scans with those sites).
The description of “locking” in ’873 patent describes what it means to
lock an image at column 2, lines 40–49, reproduced below:
In embodiments, to prevent the data from the additional
sets of intraoral images from degrading a quality of the first tooth
in the 3D model, the first set of images is automatically locked
after the first set of intraoral images is created. Additionally,
portions of the first set of intraoral images that depict the first
tooth may be exclusively used for the generation of the 3D model
of that first tooth. Thus, the additional sets of intraoral images
do not alter or add noise to a region of the 3D model depicting
the first tooth as a result of the first set of intraoral image being
locked.
Ex. 1001, 2:40–49. Claim 1 explicitly recites the steps of receiving an
intraoral image of a first intraoral site and determining the identity of that
intraoral site. The claimed locking of the intraoral image refers to the
received image of the first intraoral site. There is no basis for inferring into
the term “locking” any other limitation proposed by Patent Owner. Locking
as described in the Specification exists to prevent the data from additional
sets of intraoral images from degrading the quality of the locked image. Ex.
1001, 2:40–43. We find that no further construction is needed and apply
the same construction as applied in our Decision to Institute, i.e., we
IPR2019-00148
Patent 9,451,873 B1
16
construe “locking” to mean “storing image data and disregarding any data
from other images that spatially overlap the ‘locked’ image data.”
ANALYSIS OF PRIOR ART CHALLENGES
A. Introduction
A patent claim is unpatentable under 35 U.S.C. § 103(a) if the
differences between the claimed subject matter and the prior art are such that
the subject matter, as a whole, would have been obvious at the time the
invention was made to a person having ordinary skill in the art to which said
subject matter pertains. KSR Int’l Co. v. Teleflex Inc., 550 U.S. 398, 406
(2007). The question of obviousness is resolved on the basis of underlying
factual determinations including: (1) the scope and content of the prior art;
(2) any differences between the claimed subject matter and the prior art;
(3) the level of ordinary skill in the art; and when presented (4) objective
evidence of nonobviousness. Graham v. John Deere Co., 383 U.S. 1, 17–18
(1966).
B. Claims 1–20 as obvious over Koppelman and Paley
Claim 1 is drawn to a method. Claim 11 is drawn to a non-transitory
storage medium having instructions that when executed by a processing
device perform the claimed operations. Claim 20 is drawn to a computing
device comprising a memory (designated by Petitioner as claim element
20.1) and a processing device (designated by Petitioner as claim element
20.2). As discussed further below, claim 1 recites certain steps, i.e.,
“locking the first set of intraoral images” and “selecting, based at least in
part on the identity of the apportion of the first intraoral site, one or more
portions of the set of intraoral images depicting the preparation tooth” are
algorithmically performed. Independent claims 11 and 20 recite
corresponding steps are performed “without user input.” The remaining
IPR2019-00148
Patent 9,451,873 B1
17
limitations of claims 1, 11, and 20 are similar. Therefore, we discuss the
individual limitations of claim 1 as exemplary of the subject matter of claims
1, 11 and 20.
C. Kopelman
Kopelman discloses computer-based methods for creating and
interacting with three dimensional virtual models that are useful in dentistry,
prosthodontics, and orthodontics. Ex. 1003, 1 (title and lines 2–4).
Kopelman recognizes that when scanning an intraoral cavity to obtain 3D
data of a tooth preparation and finish line, part of the finish line, the
shoulder, other parts of the preparation, and surrounding portions of the oral
cavity may be obscured. 2:3–16. Kopelman seeks to avoid rescanning an
entire physical item when part of the physical item to be modeled is
obscured. Id. at 2:17–24. Kopelman discloses three aspects of an invention
to address these issues when generating such virtual 3D models. Id. at 5:12–
22:15.
A first aspect of Kopelman’s invention is a computer based method
for modifying a virtual model of a physical structure; the method includes:
(i) displaying an image of the virtual model, (ii) identifying at least a portion
of the virtual model to be modified by interacting with the displayed image,
(iii) modifying the virtual model by replacing in the computer system at least
a portion of the virtual model with additional 3D data obtained from the
physical structure. Ex. 1003, 5:12–20.
A second aspect of Kopelman is a computer based method that (i)
provides the computer system with a virtual model obtained from the
physical structure, (ii) provides the computer system additional 3D data
obtained from at least part of the physical structure, (iii) identifies at least a
portion of the virtual model that is to be modified with the additional 3D
IPR2019-00148
Patent 9,451,873 B1
18
data, and (iv) modifies the virtual model by replacing the identified portion
with at least part of the 3D data. Id. at 10:10–20.
A third aspect of Kopelman also concerns a computer based method
and is similar to the second aspect, except that the additional 3D data
includes at least one portion that corresponds to, but is different from, a
corresponding portion of the virtual model. Id. at 15:25–16:4. In the third
aspect, the virtual model is modified by replacing the corresponding portion
of the virtual model with the portion of the additional 3D data. Id. at 16:5–7.
Kopelman describes an embodiment in which a first virtual model
VM1 of a physical part RM1 of an intraoral cavity includes a portion DVM1
that is removed because it is not suitable, e.g., because it is obscured,
resulting in model VM1´. Ex. 1003, 22:17–23:5 (steps (a)–(e)). A second
model VM2 “representing the part DRM1 of the physical dental surface plus
additional identifying surface data ID” is developed and manipulated to
register to VM1´ obtained by rescanning DRM1 and an additional part P2,
i.e., the real dental surface surrounding DRM1, to obtain additional
identifying surface data ID. Id. at 23:6–11 (steps (f)–(g)). Using the
identifying surface data ID of second virtual model VM2, VM2 is
manipulated to register it onto modified first virtual model VM1´ so that
VM2 fits into and corresponds to the deleted model portion DVM1. Id. at
23:12–19 (steps (h)–(j)). Part DVM2 is then stitched to modified first virtual
model VM1´ to create further modified first virtual model VM1´´ and the
remainder of VM2, including the identifying surface data ID is discarded.
Id. at 23:19–24 (steps (j–k)).
In the embodiment shown in Figures 4–6, DVM1 is “marked,” i.e.,
identified by the user to demarcate an area on displayed image DI1 (step
430). Ex. 1003, 32:27–31. DVM1 is then deleted (step 440) to form VM1´
IPR2019-00148
Patent 9,451,873 B1
19
and second virtual model VM2, comprising virtual model part DVM2 of
second part RM2 of the physical dental structure and additional part P2 of
the real dental surface surrounding DRM1 to obtain additional identifying
3D surface data ID, is created. Id. at 33:3–34:29. Second virtual model
VM2 is spatially registered to VM1´ to provide composite model VM3 using
identifying surface data ID of VM2, such that DVM2 is stitched to VM1´
and the identifying surface data ID is discarded. Id. at 35:11–25. Kopelman
notes that instead of using partially overlapping data, physical parts RM1
and RM2 can be referenced to the same global coordinate system via an
optical marker whose coordinates are known and that is scanned together
with both RM2 and RM1. Id. at 35:26–30.
D. Paley
Petitioner asserts that Paley discloses creating composite 3D models
by combining intraoral images, i.e., an assortment of 2D intraoral images.
See, e.g., Pet. 33, 36, 40, 41, 42, 43, 44. Paley defines an “image” as
generally referring to “a two-dimensional set of pixels forming a two-
dimensional view of a subject within an image plane” and an “image set” as
“a set of related two-dimensional images that might be resolved into three-
dimensional data.” Ex. 1004 ¶ 22. Paley defines a “point cloud” as “a three-
dimensional set of points forming a three-dimensional view of the subject, as
might be constructed from a number of two-dimensional views.” Id. Paley
discloses using a scanner (id. ¶¶ 28, 29) to capture images of a subject, such
as a tooth (id. ¶ 32) and navigation controls to permit a user to view,
manipulate, and evaluate an acquired 3D surface representation, including
inspecting occluded features, and to permit the user to return to data
acquisition mode (id. ¶¶ 42, 58–60). Paley states:
IPR2019-00148
Patent 9,451,873 B1
20
In an example of general operation of a three-dimensional image
capture system 100, the scanner 102 may acquire two-
dimensional image sets at a video rate while the scanner 102 is
passed over a surface of the subject. The two-dimensional image
sets may be forwarded to the computer 108 for derivation of
three-dimensional point clouds. The three-dimensional data for
each newly acquired two-dimensional image set may be derived
and fitted or “stitched” to existing three-dimensional data using
a number of different techniques.
Id. ¶ 36.
Recognizing that in continuous or incremental data acquisition
systems the fitting or stitch between two frames may fail, Paley provides a
recover mode in which the system reacquires the previous scan by test fitting
a new scan to previously acquired data. Id. ¶ 38. Paley also discloses a
“landing mode” in which a user may initiate a new scan registered or
connected to an existing three-dimensional model. Id.
E. Claim 1
1. Claim Limitation 1.1
Petitioner identifies as claim element 1.1 the limitation that recites
“receiving an intraoral image of a first intraoral site.” Pet. 25. Petitioner
cites Kopelman’s first 3D virtual model VM1, obtained by scanning and
displayed as first display image DI1, as disclosing a composite image of a
physical part RM1 of the intraoral cavity. Id. at 26–27 (citing Ex. 1001,
22:18–22; 27:16–30, 28:1–31:22, 31:23–34, Fig. 3; Ex. 1024, Saber Decl. ¶¶
74–75). Petitioner cites Paley as teaching “virtual models [such as that
disclosed by Kopelman] are created by receiving intraoral images obtained
by scanning a patient’s intraoral cavity.” Id. at 28–29 (noting that Paley
discloses 2D intraoral images obtained by scanning directly from a patient’s
IPR2019-00148
Patent 9,451,873 B1
21
intraoral cavity are used to generate a 3D model (citing Ex. 1004 ¶¶ 22, 29;
Ex. 1024, Saber Decl. ¶¶ 77–79)).
The Patent Owner Response does not dispute that the combination of
Kopelman and Paley teaches receiving an image of an intraoral site. We
find that Petitioner has demonstrated that the combination of Kopelman and
Paley discloses claim limitation 1.1.
2. Claim Limitation 1.2
Petitioner identifies as claim element 1.2 the limitation that recites
“determining an identity of the first intraoral site.” Pet. 29. Noting that this
claim element encompasses making the determination manually or
automatically, Petitioner cites Kopelman’s disclosure that first virtual model
VM1 can be checked visually, and that in some embodiments, the analysis
may be carried out by a computer. Id. at 29–30 (citing Ex. 1003, 31:30–31,
22:24–25, 32:6–7; Ex. 1024, Saber Decl. ¶ 83 (manually determining the
identity), Ex. 1003, 21:27–22:3; Ex. 1024, Saber Decl. ¶ 84 (automatically
determining the identity)).
Referring to this limitation as limitation [X.2], Patent Owner notes
that Petitioner argues a user visually inspecting Kopelman’s first virtual
model performs this function because the user knows in his or her mind the
identity of the site the user is looking at. PO Resp. 26. Patent Owner also
notes that Kopelman provides controls to enable user’s viewing of the first
virtual model VM1 via display image DI 1 to determine and mark any
unacceptable part DVM1 of the model. PO Resp. 27 (citing Ex. 1003,
31:18–25, 31:31–32:31). Patent Owner argues the user’s visual inspection
and marking of model VM1 does not teach “a designation of a tooth or
portion of a jaw that specifies a location in the intraoral space” because
undesirable part DVM1 is not a designation of a physical structure. Id. at
IPR2019-00148
Patent 9,451,873 B1
22
28. The ’873 patent, however, encompasses the same approach as Patent
Owner ascribes to Kopelman. See Ex. 1001, 13:34–37 (“Processing logic
may direct the practitioner as to which intraoral site (e.g., which tooth) is to
be scanned or the dental practitioner may indicate which intraoral site is to
be scanned or has been scanned.”).
Patent Owner also notes Petitioner’s argument that the claimed
identifying step is taught by Kopelman’s disclosure of automatically
determining whether a portion of the first virtual model VM1 complies with
a predetermined requirement. PO Resp. 26, 28–30 (citing Pet. 30). Patent
Owner argues identifying an undesirable part DVM1 of the first virtual
model VM1 or determining the suitability of a physical structure RM1 for
the purpose of allowing seating of a prosthesis automatically does not teach
or suggest determining a “designation of a tooth of portion of a jaw that
specifies a location in the intraoral space.” PO Resp. 29–30 (citing
Ex. 2003, Bajaj Decl. ¶¶ 94–95, Ex. 2025, ¶¶ 98–101).
Petitioner contends we should reject Patent Owner’s arguments
because they are based on an incorrect construction of the “identity term.”
Pet. Reply 15. According to Petitioner, Kopelmen discloses this limitation
because the “first virtual model may be checked manually in some
embodiments, and automatically by computer analysis in others.” Id. at 12
(citing Pet. 29–30). Patent Owner acknowledges that either the user, by
inputting the identity of the tooth just scanned or to be scanned, or the
computer, by prompting the human to scan a particular tooth, may initiate
the interaction, but argues “the computer must determine the identity of each
intraoral site, such that it can lock the scanned image to that site” and “[t]hat
identity must be some sort of designation unique to that tooth (or jaw
portion) so that the computer can differentiate the locked image set from all
IPR2019-00148
Patent 9,451,873 B1
23
of the locked image sets that will ultimately be compiled to generate the
model.” PO Surreply 3–4.
In our discussion of the construction of the identifying term we noted
that any form or format of identification that uniquely designates a tooth or
portion of a jaw in the intraoral space satisfies the identification requirement.
See Section V.1 above. As Petitioner notes, Kopelman discloses that the
analysis of a model may be visual by the user or carried out by the computer
system. Pet 29–30 (citing Ex. 1003, 21:22–22:3). Kopelman states
explicitly “the analysis may be configured for identifying regions of the first
physical part (i.e., on the first 3D virtual model) that require physical
change.” Ex. 1003, 22:3–4. Kopelman also states “[t]he analysis software
may further identify these regions and mark them automatically” and “the
deleting/removing/replacing step can also be carried our automatically once
these regions are marked.” Id. at 22:8–10. “Thereafter the user can rework
the physical part, in particular the preparation in the regions thereof
previously identified by the analysis software and now erased and missing in
the modified first 3D virtual model.” Id. at 22:11–13. Referring to the
“corresponding part DRM1 of the physical dental surface . . . obscured by
another element” (id. at 23:1–2) Kopelman also discloses “[a] second virtual
model VM2 is created, representing the part DRM1 of the physical dental
surface plus additional identifying surface data ID” (id. at 23:6–7).
We further note Kopelman’s disclosure that physical parts, and hence
the images used to generated 3D models of such physical parts, can be
referenced to the same global coordinate system. Id. at 35:26–31.
In consideration of the above, we find Kopelman discloses several
ways of determining the identity of the intraoral site, as recited in claim
limitation 1.2. For example, Kopelman discloses that surface data (surface
IPR2019-00148
Patent 9,451,873 B1
24
data ID) can be used as a designation to identify an intraoral site. See, e.g.,
Ex. 1003, 23:6–7). Kopelman also discloses that the computer’s system
analysis software can be configured to identify a region of the first physical
part, i.e., an intraoral site, whose modeling requires change, and designate
that site by marking it automatically. See, e.g., id. at 22:10–11. Reference to
such marking is another way of determining an identity of the first intraoral
site. Finally, Kopelman discloses the parts of a model can be referenced to a
global coordinate system to designate an intraoral site from which its
identity can be determined. See, e.g., id. at 35:26–31. Thus, we are
persuaded by Petitioner’s argument that Kopelman discloses “determining
an identity of the first intraoral site” as recited in claim limitation 1.2.
3. Claim Limitation 1.3
Petitioner identifies as claim element 1.3 the limitation that recites
“algorithmically performing the following by a processing device.” Pet. 30.
Petitioner cites Kopelman as disclosing the use of a suitable computer for
processing pre-programmed algorithms. Id. at 31 (citing Ex. 1003, 31:18–
22, 35:11–25, Fig. 2; Ex. 1024, Saber Decl. ¶¶ 87–89).
Petitioner cites Kopelman as disclosing the “algorithmically
performed” locking, selecting and generating steps in claim elements 1.4–
1.6, because they can be performed automatically without user input. Id.
Petitioner specifically cites one of Kopelman’s alternative embodiments in
which step 460´ shown in Figure 9 (causing the computer system to modify
the first 3D virtual model, etc.) can be implemented automatically. Id.; Ex.
1003, Fig. 9. Petitioner notes Kopelman discloses computer system 32 is
configured to replace automatically part of the modified first 3D virtual
model VM1´ with part of second 3D virtual model VM2 and argues this
replacement involves the locking, selecting, and generating steps in claim
IPR2019-00148
Patent 9,451,873 B1
25
elements 1.4, 1.5, and 1.6. Pet. 31–32 (citing Ex. 1003, 38:30–21, 39:7–21;
Ex. 1024, Saber Decl. ¶¶ 91–92).
Patent Owner does not contend that Kopelman does not perform steps
using an algorithm; Patent Owner contends that Koppelman or the
combination of Koppelman and Paley does not teach the claimed locking,
selecting and model generating steps. PO Resp. 25–26.
In consideration of the above, we are persuaded that Petitioner has
demonstrated the combination of Kopelman and Paley discloses claim
limitation 1.3
4. Claim Limitation 1.4
Petitioner identifies as claim limitation 1.4 the limitation that recites
“locking the intraoral image.” Petitioner explains that in Kopelman
modified 3D virtual model VM1´ is locked, as in claim element 1.4, because
its data is not changed; instead areas of second virtual model VM2 that
overlap the modified first 3D virtual model are discarded. Pet. 33 (citing Ex.
1003, 35:20–25, 23:18–22, 38:24–26). As evidence of locking, Petitioner
cites Kopelman’s teaching that “part DVM2 is then stitched to modified first
virtual model VM1' in a virtual manner to create a further modified first
virtual model, i.e., composite third 3D virtual model VM3. The remainder
of the second virtual model VM2, including the identifying surface data ID
may then be discarded.” Id. (citing Ex. 1003, 35:20–25); see also id. at
23:18–22, 38:24–26 (cited by Petitioner as discussing discarding any portion
of second virtual model VM2 that overlaps modified virtual model VM1´).
Petitioner also cites the testimony of Dr. Saber and his annotated versions of
Figures 5 and 6 of Kopelman as illustrating re-scanning of second part RM2
to generate second virtual model VM2 (Ex. 1024 ¶ 98), the insertion of VM2
IPR2019-00148
Patent 9,451,873 B1
26
into VM1´ (id. ¶ 97), and the discarding of data from overlapping areas (e.g.,
the identifying surface data ID) (id. ¶ 99).
Referring to this limitation as [X.4], Patent Owner contends that
Kopelman in view of Paley fails to teach the limitation “because (1) the
combination does not [] teach or suggest an identity by which to lock image
data under the proper construction of ‘locking,’ and (2) does not disregard
all data that overlaps the alleged ‘locked image data’ (in accordance with
Petition’s proposed construction).” PO Resp. 31 (citing Ex. 2003, Bajaj
Decl. ¶¶ 97–114), PO Surreply 11. According to Patent Owner, Kopelman
does not teach or suggest an identity, nor does it associate image or image
set to an identity of an intraoral site, so there can be no “locking” of the
intraoral image. Id. at 33.
Patent Owner argues that Kopelman does not teach any particular
reason or methodology for discarding the portion of virtual model VM2 that
overlaps modified virtual model VM1´ during the creation of composite
third virtual model VM3. PO Surreply 11. Nevertheless Patent Owner
acknowledges that Kopelman states the overlapping data may be discarded.
Id. (citing Pet. 31; Ex. 1003, 23:21–22). At the same time, keying on
Petitioner’s proposed construction of “locking,” Patent Owner argues that
“in the Kopelman-Paley model data overlapping the first virtual model
VM1´ is not disregarded, but instead is used to register the second virtual
model VM2 when generating the composite virtual model VM3.” Id. at 12.
We agree that Kopelman does not disregard the overlapping data
entirely, but because Kopelman does not use insert overlapping data into
model VM1´, VM1´is “locked” as that term is used in the ’873 patent. In
some embodiments, Kopelman marks a deleted portion of a model (Ex.
1003, 32:27–34:11), rescans the physical subject matter corresponding to the
IPR2019-00148
Patent 9,451,873 B1
27
deleted part to generate a second virtual model VM2 that includes subject
matter that overlaps portions of VM1´ (id. at 34:20–35:20) and uses the
overlapping data to establish the identity of the intraoral site where VM2 is
to be inserted into VM1´ (id. at 35:17–24). After completing registration of
VM2 to VM1´, the non-overlapping portion of VM2 is inserted into VM1´
resulting in VM3. Id. Kopelman teaches the overlapping data in VM2 then
may be discarded because the overlapping data is no longer needed for
model VM3. Id. at 35:24–25. Thus, there is no alteration of the data from
VM1 remaining in VM1´ generated by removing the unusable part DVM1
from VM1. In Kopelman VM1´ is locked, as in the ’837 patent, because
“the additional sets of intraoral images do not alter or add noise to a region
of the 3D model depicting the first tooth.” See Ex. 1001, 2:46–49.
In another embodiment Kopelman states explicitly that in its computer
software “the analysis may also be configured for identifying regions of the
first physical part (i.e., on the first 3D virtual model) that require physical
change.” Ex. 1003, 22:3–4. Kopelman further teaches that model VM2 is
generated by rescanning the physical part corresponding to the part of model
DVM1 deleted to form VM1´ plus additional identifying data (surface data
ID). Id. at 23:6–11. Kopelman further teaches that VM2 is fit to VM1´ and
that the part of VM2 corresponding to the additional data used for
identifying the registration of the new scan data is discarded in VM1´´. Id. at
23:12–24. Image data that forms VM1´ is locked because the data in VM1´
is not altered. Instead, after image data VM2 is inserted successfully into
VM1´ using the overlapping data to identify the site for the insertion, the
overlapping image data from VM2 is discarded. Id. at 23:21–22. Kopelman
also discloses that physical parts, and hence the images used to generate 3D
models of such physical parts, can be referenced to the same global
IPR2019-00148
Patent 9,451,873 B1
28
coordinate system. Id. at 35:26–30. Thus, in Kopelman physical parts can
be identified and only those portions of the model that correspond to the
rescanned portion of the physical part are inserted into the original model,
the remainder of which remains locked. Id. at 35:26–30.
In consideration of the above, we are persuaded Petitioner has
demonstrated that Paley and Kopelman teaches locking the intraoral image
as recited in claim limitation 1.4.
5. Claim Limitation 1.5
Claim element 1.5 recites “selecting, based at least in part on the
identity of the first intraoral site, a portion of the intraoral image depicting a
portion of the first intraoral site.” Petitioner notes that Kopelman discloses
different requirements may relate to the identity of the first intraoral site
(e.g., the surface of interest that is configured for enabling a prosthesis to be
mounted or an abutment tooth with an inadequately defined finish line). Pet.
36 (citing Ex. 1003, 8:26–31; Ex. 1024, Saber Dec. ¶ 106), According to
Petitioner, the selection of the first modified model VM1´ (performed by
removing the deficient portion DVM1 of first virtual model VM1) is based,
at least in part, on the identity of the first intraoral site. Id. (citing Ex. 1003,
8:26–31; Ex. 1024 Saber Decl. ¶ 107). Petitioner also notes that Paley
discloses composite 3D models are based on a set of intraoral images. Id.
(citing Ex. 1004 ¶¶ 22, 29; Ex. 1024, Saber Decl. ¶ 108).
Patent Owner contends that “[b]ecause the combination of Kopelman
and Paley does not teach or suggest the claimed ‘identity,’ the combination
of Kopelman and Paley does not teach or suggest claim 1’s ‘selecting’ based
at least in part upon the identity of the first intraoral site.” PO Resp. 40.
Patent Owner argues that Kopelman’s description of the reasons for
selecting an intraoral site, e.g., failure to meet a predetermined requirement
IPR2019-00148
Patent 9,451,873 B1
29
because of the presence of an undesirable site DVM1 in first virtual model
VM1, is different from identifying the first intraoral site. PO Resp. 41–43.
According to Patent Owner “Kopelman’s deletion of the undesirable DVM1
does not disclose selecting the modified virtual model VM1´ (i.e., the
alleged portion of the intraoral image),” but instead by deleting DVM1
Kopelman “discloses selecting everything except the modified first virtual
model VM1´.” Id. at 45 (citing Ex. 2003, Bajaj Decl. ¶ 125). Patent
Owner’s argument that Kopelman teaches selecting the undesirable part to
be removed is not persuasive because it is inconsistent with Kopelman’s
disclosure of registering VM2 to VM1´ and, after completing that
registration inserting the non-overlapping parts of VM2 into VM1´, and
discarding the overlapping parts. See Ex. 1003, 35:11–31.
Patent Owner contends that, “[a]t the most fundamental level, the
claims of the ’873 patent recite a technology that prospectively addresses the
difficulty and time-sensitivity of scanning prepared finish lines by building a
virtual model using only images of properly prepared teeth.” PO Resp. 43.
According to Patent Owner, in contrast to the ’873 patent, Kopelman
“describes a technology that provides after-the-fact solutions when a scan
problem occurs or when a patient’s teeth require additional modification.”
Id. at 43–44 (citing Ex. 2003 Bajaj Decl. ¶¶ 52–53, 122). Patent Owner’s
focus on Kopelman’s “after-the-fact” solution does not recognize that Paley
teaches scanning images, developing three-dimensional data, and stitching
them together to compile a model. Ex. 1004 ¶ 36.
Petitioner contends that Patent Owner’s arguments are grounded in its
incorrect construction of “determining the identity,” as recited earlier in
claim 1. Pet. Reply 20. Petitioner also points out that the claims are not
IPR2019-00148
Patent 9,451,873 B1
30
limited to “prospective” modeling, and do not preclude correcting images of
scanned intraoral sites. Id. at 20–21.
We agree with Petitioner. The premise of Patent Owner’s argument is
that Kopelman does not teach selecting a portion of the intraoral image
based on the identity of the intraoral site because Kopelman does not teach
determining the identity of the intraoral site. PO Resp. 40. Patent Owner’s
related arguments that Petitioner relies on Kopelman’s description of a
predetermined requirement for the claimed identity (PO Resp. 41–43) are
also not persuasive. Although satisfying a predetermined requirement may
be the reason for modifying a model, Kopelman discloses several ways to
designate the locations of the model to be modified and to remain
unmodified. Patent Owner’s arguments are negated in our discussion above
finding that Petitioner has demonstrated the combination of Paley and
Kopelman teach the claimed step of determining the identity of an intraoral
site in a number of ways. See Section VI.E.1 above.
In consideration of the above, we find Petitioner has demonstrated
that the combination of Kopelman and Paley discloses “selecting, based at
least in part on the identity of the first intraoral site, a portion of the intraoral
image depicting a portion of the first intraoral site” as recited in claim
limitation 1.5.
6. Claim Limitation 1.6
Petitioner designates as claim element 1.6, the limitation that recites
“generating a model comprising the first intraoral site based at least in part
on the locked intraoral image, wherein the portion of the locked intraoral
image is used for a first region of the model, and wherein data from one or
more additional intraoral images that also depict the portion of the first
intraoral site is not used for the first region of the model.” Pet. 37. As to
IPR2019-00148
Patent 9,451,873 B1
31
claim element 1.6, Petitioner cites Kopelman’s description of generating
virtual model VM3 by stitching part DVM2 from second virtual model VM2
to modified first virtual model VM1´. Id. (citing Ex. 1003, 35:11–15).
Petitioner relies on the following disclosure in Kopelman:
In step 460, and referring also to Fig. 6, the second 3D virtual
model VM2 is spatially registered with respect to the modified
first 3D virtual model VMl´ to provide a composite third 3D
virtual model VM3, wherein the part DVMl that was previously
deleted/removed is at least partially replaced with a
corresponding part the second 3D virtual model VM2. In
particular, the second virtual model VM2 is manipulated in the
computer system 32 to register the second virtual model VM2
onto the modified first virtual model VMl´. In this connection,
the identifying surface data ID of second virtual model VM2 may
be useful as it may be aligned with corresponding parts of the
modified first virtual model VMl´, since the surface data for part
P2 of the real physical dental surface should be nominally
identical in both scans. In this aligned position, part DVM2 of
the second virtual model VM2 fits in and corresponds to at least
a portion of the deleted portion DVMl, and part DVM2 is then
stitched to modified first virtual model VMl´ in a virtual manner
to create a further modified first virtual model, i.e., composite
third 3D virtual model VM3. The remainder of the second virtual
model VM2, including the identifying surface data ID may then
be discarded.
Ex. 1003, 35:11–25 (emphases omitted).
Petitioner notes that in this process some of the second virtual model
VM2 (e.g., the surface data ID) is discarded and first virtual model VM1´ is
used without modification (i.e., it remains locked), so the first portion of the
first intraoral site is used without modification and the data from VM2 that
overlaps the locked portion of the first intraoral site is discarded. Pet. 40
(citing Ex. 1024, Saber Decl. ¶ 113).
IPR2019-00148
Patent 9,451,873 B1
32
Patent Owner argues that because Kopelman and Paley do not teach
or suggest “locking [of] the intraoal image” to generate a locked intraoral
image, the references do not teach or suggest generating a model based in
part on the locked intraoral image. PO Resp. 46. As we discuss in Section
1.4 above, we are not persuaded by Patent Owner’s argument that
Kopelaman and Paley fail to disclose the locking limitation. As we are
persuaded that Paley and Kopelman teach the claimed locking of the
intraoral image, e.g., VM1´ discussed above, we find the subject matter
Petitioner cites in Kopleman discloses claim limitation 1.6.
Claim element 1.6 refers to the locked intraoral image, as
distinguished from the locked portion of a model. PO Resp. 47–49. We
address patent Owner’s arguments concerning the distinction between
images and models in our discussion of reasons to combine Kopelman and
Paley in Section VI.F below.
F. Reasons to combine Teachings of Paley and Kopelman
Petitioner contends a person of ordinary skill would have recognized
from Paley that a 3D model could be generated from a composite of two
dimensional models or from processing a three dimensional set of points
(Paley’s point cloud data) into a 3D virtual model. Id. at 76 (citing Ex.
1004, ¶¶ 22, 28–29). Petitioner further contends that a person of ordinary
skill would have appreciated that Kopelman’s virtual models may be
generated by compositing scanned intraoral data in the manner disclosed n
Paley. Id. (citing Ex. 1003 31:20-22 (discussing turning raw data into a
model); Ex.1024, ¶¶29, 261 (discussing that it was well-known to produce
3D virtual models by compositing 2D images); Ex. 1007, 1:42–50
(discussing combining intraoral images to generate a three-dimensional
IPR2019-00148
Patent 9,451,873 B1
33
model); Ex.1008, 7:51-62 (discussing capturing multiple images and sending
to an image processor for generation of a three-dimensional model)).
Petitioner acknowledges that Kopelman does not disclose a scanned
image obtained directly from the intraoral cavity of a patient. Pet. 75.
According to Petitioner, a person of ordinary skill would have understood
the modification substitutes one known element (intraoral images for
compiling a virtual model) for another (a virtual model from intraoral
images) because a set of uncombined intraoral images is precursor data of
the virtual model in which the images are combined and because the results
are predictable. Id. at 76–77. Petitioner contends that a person of ordinary
skill would have “readily appreciated that the locking, selecting, and
generation of Kopelman would have been obvious to perform on Paley’s
image set.” Id. at 77. According to Petitioner a person of ordinary skill
would have been motivated to combine these teachings because performing
the processing before combining the intraoral images into a composite
virtual model is more efficient, i.e., “the intraoral images only need to be
combined once to create the final model as opposed to combining the images
into a first virtual model VM1, combining the additional images to create the
second virtual model VM2, and then creating a third virtual model VM3
based on the first and second models VM1, VM2.” Id. at 79 (citing
Ex. 1024, Saber Decl. ¶266).
Arguing that Kopelman provides a tool for a purpose different from
that of the ’873 patent, Patent Owner contends that Kopelman concerns
“editing models on the back end,” in contrast to the ’873 patent, which “is
about curating data and parsing data and designing certain pieces of data to
render certain portions of the model . . . all being done on the front end to
generate the perfect model from the beginning.” Hr’g Tr. 16; PO Resp. 54.
IPR2019-00148
Patent 9,451,873 B1
34
According to Patent Owner “Kopelman is about what happens after you’ve
generated a perfect model . . . that no longer accurately represents your
dental history . . . how we can go into the model or delete part of the model
you don’t like and, then build a new model in the corresponding space the
reflects the updated dental history.” Id. Patent Owner contends that
Kopelman’s purpose of correcting a model that contains an undesirable part
is not served by performing Kopelman’s method on scanned images instead
of models, because doing so would not create a composite model. PO Resp.
53.
Noting that Kopelman ends with generating a virtual model, Patent
Owner contends that Petitioner’s proposed modification of Kopelman would
cripple Kopelman’s ability to remove undesirable parts of a virtual model in
a post-processing step and degrade performance by adding previously
unperformed suitability processing operations. PO Resp. 55. Patent Owner
also contends Petitioner fails to explain how a person of ordinary skill would
have combined the teachings of Kopelman and Paley and, instead, relies on
improper hindsight to rearrange Kopelman’s steps without explaining how
using Paley’s set of intraoral images improves efficiency. Id. at 57–58.
Patent Owner also argues that Petitioner’s proposed combination of
references frustrates refining a previously generated model. PO Resp. 54–
55. Petitioner responds that Patent Owner’s argument fails to acknowledge
that Petitioner’s proposed combination of Kopelman with Paley performs
Kopelman’s steps at the image level, such that the model is refined at that
source image level before being produced. Pet. Reply. 25. As such, there
are no undesirable parts that would be necessary to remove. Id.
Patent Owner further argues that Kopelman’s suitability determination
could not have been performed on individual images because individual
IPR2019-00148
Patent 9,451,873 B1
35
images include less image information to measure prosthetic thickness of an
intraoral site. PO Resp. 58. Petitioner responds that Patent Owner fails to
recognize that Kopelman’s virtual model is created from intraoral images
that are the source of such information. Id. at 26.
We agree with Petitioner that a person of ordinary skill would have
recognized from Paley a reason to modify Kopelman for use in assembling
intraoral images before completing a virtual model. Paley states, for
example, that “three-dimensional data for each newly acquired two
dimensional image set may be derived and fitted or “stitched” to existing
three-dimensional data using a number of different techniques.” Ex. 1004 ¶
36. Addressing the previous state of the art, Paley explicitly states that
“[c]urrent techniques generally separate acquisition from registration with
registration being performed in a post-processing step separate from data
acquisition” and that “[w]hile this permits relatively exhaustive processing,
it poses a significant disadvantage because accuracy and completeness of the
overall scan cannot be evaluated until after the same has been completed.”
Ex. 1004 ¶ 7. Thus, Paley explicitly recognizes the efficiency considerations
identified by Petitioner.
Patent Owner acknowledges that Paley teaches models built out of
scanned data, but does not mention Paley’s disclosure of a “recover mode”
that can be entered upon detecting a failure of the fit or stitch to a previous
frame as the model is being constructed. Ex. 1004 ¶ 38 (describing
reacquisition of a continuous or incremental scan by test fitting new scan
data to previously acquired data). Paley also discloses a related landing
mode in which “a user may attempt to initiate a new scan registered or
connected to an existing three-dimensional model.” Id. Thus, Paley teaches
an ordinarily skilled artisan to stitch image sets together to build a model or
IPR2019-00148
Patent 9,451,873 B1
36
modify an existing model. As Kopelman teaches a more efficient approach
to modifying existing models and Paley teaches techniques for use with
images and models, we are persuaded that Petitioner has demonstrated a
person of ordinary skill would have been motivated to combine the teachings
of Kopelman and Paley to arrive at the limitations recited in claim 1.
G. Independent claims 11 and 20
Independent claim 11 is drawn to a non-transitory storage medium
having instructions that cause a processing device to perform the operations
recited in the claim limitations. The operations recited in claim 11 are
similar to the method steps recited in claim 1, but include receiving a first set
of oral images of a preparation tooth showing a finish line of the preparation
tooth and corresponding limitations on the locking, selecting, and generating
operations. Independent claim 20 is drawn to a computing device
comprising a memory and a computing device to perform operations similar
to those recited in claim 1. Claims 11 and 20 explicitly recite that the
locking and selecting operations are performed without user input. None of
the differences in claims 11 and 20 from the limitations in claim 1 have an
effect on our analysis of Petitioner’s challenge based on Kopelman and
Paley and no such arguments are before us.
In consideration of the above, we are persuaded that the limitations of
claims 11 and 20 are disclosed by Kopelman and Paley for the same reasons
we find these references disclose the limitations of claim 1.
H. Claim 2
Claim 2 depends from claim 1. Identifying as claim element 2.1 the
limitation that recites “the intraoral image is a set of intraoral images, ”
Petitioner cites Paley ’112 as disclosing generating a 3D virtual model by
IPR2019-00148
Patent 9,451,873 B1
37
compositing a set of intraoral images. Pet. 41 (citing Ex. 1004 ¶¶ 22, 29; Ex.
1024, Saber Decl. ¶ 118).
Petitioner identifies as claim element 2.2 the limitation that recites
“locking the first set of intraoral images.” Id. Petitioner cites Kopelman as
disclosing locking first virtual model VM1´ and Paley as disclosing
generating a 3D virtual model by combining intraoral images. Id. (citing Ex.
1024, Saber Decl. ¶¶ 120–121).
Petitioner identifies as claim element 2.3 the limitation that recites
“for each intraoral image that is a member of the first set of intraoral images,
selecting a portion of that intraoral image depicting the portion of the
intraoral site.” Id. Petitioner cites Kopelman as disclosing selecting a
portion of first virtual model VM1 depicting a portion of the intraoral site
and Paley as disclosing generating a 3D model by combining a set of
intraoral images. Id. at 41–42. Patent Owner does not respond explicitly to
Petitioner’s contentions concerning the additional limitations recited in
claim 2.
Patent Owner does not respond directly to Petitioner’s contentions
concerning the additional limitations recited in claim 2. In consideration of
the above, we are persuaded that Petitioner has demonstrated a person of
ordinary skill would have had reason to combine the teachings of Kopelman
and Paley to arrive at the limitations of these claims.
I. Claims 3 and 12
Claims 3 and 12 depend from claims 2 and 11, respectively, and recite
similar limitations. For purposes of this Decision, we treat claim 3 as
representative. Petitioner identifies as claim element 3.1 the limitation that
recites “receiving a second set of intraoral images of a second intraoral site,
wherein one or more portions of the second set of intraoral images also
IPR2019-00148
Patent 9,451,873 B1
38
depict the first intraoral site.” Pet. 42. Petitioner contends that the “first
intraoral site” is commensurate to the preparation tooth shown in Figure 4 of
Kopelman and that “the second virtual model VM2 (taken of a ‘second
intraoral site’ includes a portion (identifying surface data ID) that also
depicts the first intraoral site.” See id. at 42–43.
Petitioner identifies as claim element 3.2 the limitation that recites
“disregarding the one or more portions of the second set of intraoral images
when generating the model, wherein the second set of intraoral images does
not alter or add noise to the first region of the model as a result of the first
set of intraoral images being locked.” See, id. at 43–44. Petitioner cites
Kopelman as disclosing that portions of the second virtual model VM2 are
disregarded and Paley as disclosing that a virtual model is created by
compiling an assortment of 2D images. See id. at 43–44.
Patent Owner does not respond directly to Petitioner’s contentions
concerning the additional limitations recited in claims 3 and 12. In
consideration of the above, we are persuaded that Petitioner has
demonstrated a person of ordinary skill would have had reason to combine
the teachings of Kopelman and Paley to arrive at the limitations of these
claims.
J. Claims 4 and 13
Claims 4 and 13 depend from claims 3 and 12, respectively and recite
similar limitations. For purposes of this Decision, we treat claim 4 as
representative. Petitioner identifies as claim element 4.1 the limitation that
recites “determining an identity of the second intraoral site.” Pet. 44. As
discussed above, Petitioner identifies Kopelman’s preparation tooth of the
first virtual model VM1´ as commensurate to the “first intraoral site.” Id.
Petitioner identifies real part RM2 depicted by second virtual model VM2 as
IPR2019-00148
Patent 9,451,873 B1
39
the claimed “second intraoral site.” Id. at 45 (citing Ex. 1024, Saber Decl.
¶ 137). Petitioner argues that in Kopelman the identity of the second
intraoral site is determined by computer system 32 or visually determined
when a user sees a defective portion DVM1 and re-scans a portion RM2 of a
patient’s cavity. Id. (citing Ex. 1003, 38:30–39:21, Fig. 5; Ex. 1024, Saber
Decl. ¶ 138).
Petitioner identifies as claim element 4.2 the limitation that recites
“locking the second set of intraoral images.” Pet. 45. Petitioner cites
Kopelman as disclosing replacement portion DVM2 of second virtual model
VM2 is spatially registered and used to replace the first portion DVM1 of
the first 3D virtual model; and the replacement portion DVM2 of second
virtual model VM2 is “locked” because the data is used, without alteration,
and the composite third 3D virtual model VM3. Pet. at 45–46 (citing Ex.
1003, 38:30–39:21; Ex. 1024, Saber Decl. ¶ 140).
Petitioner cites Paley as evidence that a virtual model is created by
compiling an assortment of 2D intraoral images. Id. at 46 (citing Ex. 1024,
Saber Decl. ¶ 41).
Petitioner identifies as claim element 4.3 the limitation that recites
“selecting, based at least in part on the identity of the second intraoral site,
one or more portions of the second set of intraoral images depicting a
portion of the second intraoral site, wherein the model further comprises the
second intraoral site, and wherein the second intraoral site in the model is
based at least in part on the locked second set of intraoral images.” Pet. 46.
Petitioner cites Kopelman as disclosing spatially registering second virtual
model VM2 with first modified virtual model VM1´ and using the portion
DVM2 of the second virtual model to replace the defective portion DVM1
of the first virtual model. Id. (citing Ex.1003, 38:30-39:21; Ex. 1024, Saber
IPR2019-00148
Patent 9,451,873 B1
40
Decl. ¶143). Petitioner contends that as a result, the third 3D virtual model
is generated using the locked portion DVM2 of virtual model VM2. Id.
According to Petitioner, composite third 3D virtual model VM3 thus
includes a portion of the second virtual model depicting a portion of the
second intraoral site based at least on part of the locked part DVM2 of the
second virtual model VM2. Id.
Petitioner again cites Paley as disclosing a virtual model is created
based on compiling an assortment of 2D intraoral images. Id.
Patent Owner does not respond directly to Petitioner’s contentions
concerning the additional limitations recited in claims 4 and 13. In
consideration of the above, we are persuaded that Petitioner has
demonstrated a person of ordinary skill would have had reason to combine
the teachings of Kopelman and Paley to arrive at the limitations of these
claims.
K. Claims 5 and 14
Claims 5 and 14 depend from claims 4 and 13, respectively, and recite
similar limitations. For purposes of this Decision, we treat claim 5 as
representative. Petitioner identifies as claim element 5.1 the limitation that
recites “stitching the first set of images to the second set of images.”
Petitioner cites Kopelman as disclosing part DVM2 is stitched to the
modified virtual model VM1´. Pet. 47–48 (citing Ex. 1003, 38:20–24; Ex.
1024, Saber Decl. ¶148). Petitioner cites Paley as explaining a model is
generated from intraoral images. Id. at 48 (citing Ex. 1024, Saber Decl. ¶
150).
Petitioner identifies as claim element 5.2 the limitation that recites
“identifying the one or more discrepancies of overlapping data between the
first set of images and the second set of images.” Id. Petitioner cites
IPR2019-00148
Patent 9,451,873 B1
41
Kopelman as disclosing a lining the first and second virtual models to
identify discrepancies in overlapping data. Id. at 48–49 (citing Ex. 1003
36:27–37:6, 23–25; Ex. 1024, Saber Decl ¶ 151). Petitioner cites Paley as
discussed above for the proposition that a model is created by receiving 2D
intraoral images obtained by scanning the intraoral cavity. Id. at 48.
Petitioner identifies as claim element 5.3 the limitation that recites
“prioritizing the first set of images over the second set of images.” Pet. 49.
Petitioner cites Kopelman as disclosing that the remainder of the second
virtual model VM2 including the identifying surface data ID may be
discarded. Id. (citing Ex. 1003, 38:20–26). According to Petitioner this
means that the first virtual model VM1´ is prioritized over the second virtual
model VM2 where the models overlap. Id. (citing Ex. 1024, Saber Decl.
¶154). As to this element, Petitioner cites Paley as explaining a model is
generated from intraoral images. Id. (citing Ex. 1024, Saber Decl. ¶155).
Petitioner identifies as claim element 5.4 the limitation that recites
“applying a weighted average of the overlapping data between the first set of
images and the second set of images, wherein the data from the first set of
images has a higher weight than the data from the second set of images.”
Pet. 49–50. Petitioner argues that because Kopelman discloses discarding
the remainder of second virtual model VM2, modified first virtual model
VM1´ in Kopelman’s weighted higher (100%) than second virtual model
VM2 (0%). Id. at 50 (citing Ex. 1024, Saber Decl. ¶ 157). Petitioner again
cites Paley as explaining a model is generated from intraoral images. Id.
(citing Ex. 1024, ¶ 157).
Patent Owner does not respond directly to Petitioner’s contentions
concerning the additional limitations recited in claims 5 and 14. In
consideration of the above, we are persuaded that Petitioner has
IPR2019-00148
Patent 9,451,873 B1
42
demonstrated a person of ordinary skill would have had reason to combine
the teachings of Kopelman and Paley to arrive at the limitations of these
claims.
L. Claims 6 and 15
Claims 6 and 15 depend from claims 3 and 12, respectively, and recite
similar limitations. For purposes of this Decision, we treat claim 6 as
representative. Petitioner identifies as claim element 6.1 the limitation that
recites “the locked first set of intraoral images comprises a boundary for the
selected portions of the first set of intraoral images.” See Pet. 50. The
Petition includes an annotated version of Figure 4 of Kopelman to illustrate
that the “boundary” is the outer periphery of first 3D virtual model VM1
until the deleted part DVM1, such that the portion of first virtual model
VM1 that is not deleted part DVM1 is “inside” the boundary and the deleted
part is “outside” the boundary. Id. at 51. Petitioner again cites Paley as
evidence that a virtual model is created by compiling an assortment of 2-D
intraoral images. Id. at 52 (citing Ex. 1024, Saber Decl. ¶ 163).
Petitioner identifies as claim element 6.2 the limitation that recites
“wherein the one or more portions of the second set of intraoral images that
are inside of the boundary are not applied for the model.” Pet. 52.
Petitioner notes that the area or outside the inner red line within the outer
peripheral red line in the annotated version of Figure 4 of Kopelman on page
51 of the Petition is “not applied for the third 3D virtual model VM3.” Id. at
52 (citing Ex.1003, 35:11–25; Ex. 1024, ¶ 165).
Petitioner identifies as claim element 6.3 the limitation that recites
“wherein one or more additional portions of the second set of intraoral
images that are outside of the boundary are applied for the model.” Pet. 52–
53. Referencing its further annotated Figure 4 of Kopelman on page 53 of
IPR2019-00148
Patent 9,451,873 B1
43
the Petition, Petitioner notes that the portion of 3D virtual model VM2 that
is within the inner red line DVM1 and, thus outside the boundary, is applied
for the model VM3. Pet. 53. Patent Owner does not respond directly to
Petitioner’s contentions concerning the additional limitations recited in
claims 6 and 15. In consideration of the above, we are persuaded that
Petitioner has demonstrated a person of ordinary skill would have had reason
to combine the teachings of Kopelman and Paley to arrive at the limitations
of these claims.
M. Claims 7 and 16
Claims 7 and 16 depend from claims 1 and 11, respectively, and recite
similar limitations. For purposes of this Decision, we treat claim 7 as
representative. Claim 7 recites the further limitation “receiving and
indication of the identity of the intraoral site, wherein the identity of the
intraoral site is determined based at least in part on the indication.”
Petitioner cites Kopelman as disclosing that a user may mark part DVM1 of
first 3D virtual model VM1 to provide an indication of the identity of the
intraoral site. Pet. 54 (citing Ex. 1003, 22:8–11, 32:27 –31; Ex. 1024, Saber
Decl. ¶¶ 171–172).
Patent Owner does not respond directly to Petitioner’s contentions
concerning the additional limitations recited in claims 7 and 16. In
consideration of the above, we are persuaded that Petitioner has
demonstrated a person of ordinary skill would have had reason to combine
the teachings of Kopelman and Paley to arrive at the limitations of these
claims.
N. Claims 8 and 17
Claims 8 and 17 depend from claims 1 and 11, respectively, and recite
similar limitations. For purposes of this Decision, we treat claim 8 as
IPR2019-00148
Patent 9,451,873 B1
44
representative. The preamble of claim 8 recites the further feature “wherein
the locked intraoral image comprises an anomaly within the portion of the
locked intraoral image.” Petitioner cites Kopelman as disclosing the first
virtual model VM1 as an undesirable part DVM1. Pet. 54–55. Petitioner
identifies as claim element 8.1 the limitation that recites “identifying the
anomaly.” Pet. 55. Petitioner cites Kopelman as disclosing manually or
automatically identifying portion DVM1 of first virtual model VM1 as an
anomaly. Id. (citing Ex. 1003, 32:21–36).
Petitioner identifies as claim element 8.2 the limitation that recites
“determining a border of the anomaly.” Id. Petitioner cites Kopelman as
disclosing defective part DVM1 of first virtual model VM1 is identified and
removed and that the removal of part DVM1 includes determining a border
of the portion of DVM1 either manually or automatically. Id. (citing Ex.
1003, 33:3–16, 38:30–39:21; Ex. 1024 ¶ 178).
Petitioner identifies as claim element 8.3 the limitation that recites
“receiving an additional image of the first intraoral site.” Id. Petitioner cites
Kopelman as disclosing receiving a second virtual model VM2 that also
depicts the first intraoral site. Id. at 56. Petitioner also cites Paley as
discussed above for the proposition that a model is created by receiving 2D
intraoral images obtained by scanning the intraoral cavity. Id.
Petitioner identifies as claim element 8.4 the limitation that recites
“updating the model based on replacing data within the border from the
intraoral image with additional data from the additional image.” Id.
Petitioner contends that Kopelman discloses third virtual model VM3 is
created by updating first modified virtual model VM1´ to replace data from
within the border of part DVM1 with data from part DVM2 of second virtual
IPR2019-00148
Patent 9,451,873 B1
45
model VM2, and that Paley teaches that intraoral images are used to
generate virtual models. Id. (citing Ex. 1024, Saber Decl. ¶ 182).
Patent Owner does not respond directly to Petitioner’s contentions
concerning the additional limitations recited in claims 8 and 17. In
consideration of the above, we are persuaded that Petitioner has
demonstrated a person of ordinary skill would have had reason to combine
the teachings of Kopelman and Paley to arrive at the limitations of these
claims.
O. Claims 9 and 18
Claims 9 and 18 depend from claims 8 and 17, respectively, and recite
similar limitations. For purposes of this Decision, we treat claim 9 as
representative. Claim nine recites the additional limitation that “the anomaly
comprises at least one of a void in the intraoral image, noise in the intraoral
image, or unrealistic data in the intraoral image.” Petitioner cites Kopelman
as disclosing a correction operation performed on a “distorted or deficient”
area of the first 3D virtual model. Pet. 56 (citing Ex. 1003, 32:21–36).
Petitioner contends that a distorted or deficient area has a void, undesired
noise, or unrealistic data. Id. (citing Ex. 1024, Saber Decl. ¶184).
Patent Owner does not respond directly to Petitioner’s contentions
concerning the additional limitations recited in claims 9 and 18. In
consideration of the above, we are persuaded that Petitioner has
demonstrated a person of ordinary skill would have had reason to combine
the teachings of Kopelman and Paley to arrive at the limitations of these
claims.
P. Claims 10 and 19
Claims 10 and 19 depend from claims 1 and 11, respectively, and
recite similar limitations. For purposes of this Decision, we treat claim 10 as
IPR2019-00148
Patent 9,451,873 B1
46
representative. The preamble of claim 10 recites “wherein the portion of the
intraoral site comprises a preparation tooth of the intraoral site.” Petitioner
cites Kopelman as disclosing scanning an intraoral site that includes a
preparation tooth. Pet. 57.
Petitioner identifies as claim element 10.1 the limitation that recites
“determining that data for the preparation tooth is incomplete.” Id.
Petitioner cites Kopelman as disclosing that the preparation tooth surface
may be the area which has defective data. Id. (citing Ex. 1003, 8:26–31,
31:10–17).
Petitioner identifies as claim element 10.2 the limitation that recites
“receiving an additional intraoral image of the preparation tooth.” Petitioner
cites Kopelman’s second virtual model VM2 as showing the preparation
tooth when it has defective data. Id. at 57–58 (citing Ex. 1003, 8:26–9:3;
Ex. 1024, Saber Decl. ¶ 191. Petitioner cites Paley as discussed above for
the proposition that a model is created by receiving 2D intraoral images
obtained by scanning the intraoral cavity. Id.
Petitioner identifies as claim element 10.3 the limitation that recites
“identifying a border for an edge of the preparation tooth where the data for
the preparation tooth is incomplete.” Referencing an annotated version of
Kopelman Figure 4 on page 59 of the Petition, Petitioner states that
Kopelman’s replacement of defective part DVM1 of first virtual model VM1
involves identifying the border for where the defective part will be removed.
Id. at 58–59. Petitioner notes that Kopelman discloses a likely area for the
anomaly is a preparation tooth so that when the anomaly occurs Kopelman
will identify a border for the edge of the preparation tooth with the data for
the preparation tooth is incomplete. Id. at 59 (citing Ex. 1003, 8:26– 9:3;
Ex. 1024 ¶ 195).
IPR2019-00148
Patent 9,451,873 B1
47
Petitioner identifies as claim element 10.4 the limitation that recites
“updating the model based on replacing data outside the border with
additional data from the additional image to expand the model at the edge.”
Pet. 60. Referencing the annotated version of Kopelman’s Figure 4 on page
60 of the Petition, Petitioner contends that the portion of Kopelman’s virtual
model VM2 that is within the inner red line DVM1, and thus outside the
border, is applied for model VM3. According to Petitioner when additional
DVM2 data from additional image VM2 is used to update the area outside
the border, the model will be expanded to fill in data to show the edge of the
preparation tooth. Id. at 60–61 (citing Ex. 1003, 35:11– 25; Ex. 1024 ¶¶
197–198).
Patent Owner does not respond directly to Petitioner’s contentions
concerning the additional limitations recited in claims 10 and 19. In
consideration of the above, we are persuaded that Petitioner has
demonstrated a person of ordinary skill would have had reason to combine
the teachings of Kopelman and Paley to arrive at the limitations of these
claims.
Q. Patent Owner’s Declarants
Petitioner contends that Patent Owner’s computer expert Dr. Bajaj
lacks dental experience and does not qualify as a person of ordinary skill in
the art, as that term has been applied in this proceeding. Pet. Reply 27.
Petitioner further argues that Patent Owner’s dental expert Dr. Valley is not
a computer expert and did not communicate with Dr. Bajaj. Id. at 28.
Petitioner argues that by failing to present a cohesive set of declarations to
support its multi-disciplinary approach, Patent Owner has presented the
panel with contradictory declarations requiring the panel to sift through the
IPR2019-00148
Patent 9,451,873 B1
48
evidence to distinguish relevant and credentialed testimony in each
declaration. Id.
Patent Owner notes that there is no requirement Dr. Bajaj be a person
of ordinary skill, but that Dr. Bajaj has extensive experience in dental
imaging. PO Surreply 6–7. Patent Owner further states that Dr. Valley and
Dr. Bajaj deliberately did not consult, and that they independently
corroborate each other in the ultimate conclusions. Id. at 8.
Petitioner has not moved to exclude the testimony of either Dr. Valley
or Dr. Bajaj. We have considered the testimony of both witnesses in their
respective contexts and given the weight we deem appropriate to their
testimony.
CONCLUSION4
Having reviewed the evidence and arguments of record, we find that
Petitioner has demonstrated by a preponderance of the evidence that claims
1–20 of the ’873 patent are unpatentable as obvious under 35 U.S.C.
§ 103(a) over the combined teachings of Kopelman and Paley.

4 Should Patent Owner wish to pursue amendment of the challenged claims
in a reissue or reexamination proceeding subsequent to the issuance of this
decision, we draw Patent Owner’s attention to the April 2019 Notice
Regarding Options for Amendments by Patent Owner Through Reissue or
Reexamination During a Pending AIA Trial Proceeding. See 84 Fed. Reg.
16,654 (Apr. 22, 2019). If Patent Owner chooses to file a reissue application
or a request for reexamination of the challenged patent, we remind Patent
Owner of its continuing obligation to notify the Board of any such related
matters in updated mandatory notices. See 37 C.F.R. § 42.8(a)(3), (b)(2).
IPR2019-00148
Patent 9,451,873 B1
49

In summary:

Claims 35 U.S.C. § Basis
Claims
Shown
Unpatentable
Claims
Not shown
Unpatentable
1–20 103(a) Kopelman,
Paley
1–20
Overall
Outcome
103(a) Kopelman,
Paley
1–20
IPR2019-00148
Patent 9,451,873 B1
50
PETITIONER

Todd R. Walters
Roger H. Lee
Andrew R. Cheslock
Bryan Cannon
BUCHANAN INGERSOLL & ROONEY PC
todd.walters@bipc.com
roger.lee@bipc.com
andrew.cheslock@bipc.com
bryan.cannon@bipc.com

PATENT OWNER

Robert G. Sterne
Jason D. Eisenberg
Salvador M. Bezos
Kristina Caggiano Kelly
Imhotep S. Durham
STERNE, KESSLER, GOLDSTEIN & FOX P.L.L.C.
rsterne-ptab@sternekessler.com
jasone-ptab@sternekessler.com
sbezos-ptab@sternekessler.com
kckelly-PTAB@sternekessler.com
idurham-PTAB@sternekessler.com