3Shape AS v. Align Technology Inc

1. Case Identification

• Case #: IPR2019-00148
• Patent #: 9,451,873
• Filed: November 10, 2018
• Petitioner(s): 3Shape A/S and 3Shape Inc.
• Patent Owner(s): Align Technology, Inc.
• Challenged Claims: 1-20

2. Patent Overview

• Title: Processing Intraoral Images
• Brief Description: The ’873 patent relates to methods for processing intraoral images obtained by scanning a patient's mouth to generate an accurate 3D virtual model. The methods involve "locking" a first set of high-quality images (e.g., of a prepared tooth) to prevent them from being degraded by data from subsequent, potentially lower-quality, overlapping scans.

3. Grounds for Unpatentability

Ground 1: Claims 1-20 are obvious over Kopelman in view of Paley.

• Prior Art Relied Upon: Kopelman (WO 2012/011101) and Paley (Application # 2007/0172112).
• Core Argument for this Ground:
  • Prior Art Mapping: Petitioner argued that Kopelman disclosed nearly all elements of the challenged claims but within the context of manipulating a composite 3D virtual model. Kopelman taught identifying a defective portion of a first virtual model, rescanning the corresponding physical area to create a second virtual model, and then stitching the corrected data from the second model into the first while discarding overlapping data. This process of discarding overlapping data from the second scan constituted the claimed "locking." However, the ’873 patent claims operated on "intraoral images," which Petitioner construed as the individual 2D scans, not the composite 3D model. Petitioner asserted that Paley taught the conventional and well-known step of creating a 3D virtual model (like Kopelman's) by compiling a set of individual 2D intraoral images. Therefore, Paley supplied the teaching of using a set of intraoral images as the precursor data for the process disclosed in Kopelman.
  • Motivation to Combine: Petitioner presented two primary motivations for a Person of Ordinary Skill in the Art (POSITA) to combine the references. First, it would have been an obvious substitution of one known element for another—applying Kopelman’s processing steps to the precursor intraoral images (taught by Paley) instead of to the final virtual model. Since Paley taught that virtual models are simply compilations of images, a POSITA would have found it a simple and predictable design choice to perform the data selection and locking on the images themselves. Second, a POSITA would combine the teachings to improve processing efficiency. Kopelman's method required generating multiple separate virtual models. Applying the locking and selection steps directly to the uncombined image sets before generating a final model would be more efficient, as it would require only one model-generation step instead of three.
  • Expectation of Success: Petitioner contended a POSITA would have had a reasonable expectation of success because the image processing techniques described by Kopelman for manipulating a virtual model were well-known and directly applicable to manipulating the underlying set of images. The combination involved applying known techniques to a known type of data to achieve a predictable improvement in efficiency.

4. Key Claim Construction Positions

• "intraoral image" / "intraoral images": Petitioner argued for a construction of "a scanned image obtained directly from the intraoral cavity of a patient." This was central to the obviousness argument, as it distinguished the claimed invention from Kopelman's disclosure, which operated on a composite "three-dimensional virtual model." Petitioner asserted this construction created the need to combine Kopelman with Paley, which explicitly taught that such virtual models are generated from sets of individual images.
• "algorithmically performing": Petitioner proposed this term meant "automatically performing without user input." The petition cited portions of the ’873 patent specification that consistently contrasted "algorithmic" processes with alternative "manual" user-driven actions, supporting the interpretation that the term required an automated, computer-driven process.
• "locking": Petitioner proposed a construction of "storing image data and disregarding any data from other images that spatially overlap the 'locked' image data." This interpretation was based on the ’873 patent's description of preventing data from a second scan from altering or adding noise to the data from a first, higher-quality scan in overlapping regions of a 3D model.

5. Arguments Regarding Discretionary Denial

• Petitioner argued that discretionary denial under §314(a) was inappropriate for two primary reasons. First, the petition was not a "follow-on" petition, as the ’873 patent had not been previously challenged in any prior AIA trial. Second, while a parallel district court litigation existed, the ’873 patent was no longer being asserted in that case, meaning the issues in the inter partes review (IPR) were not co-extensive with the litigation and institution would not be inefficient.
• Petitioner also argued against denial under §325(d), contending that the specific combination of Kopelman and Paley was not previously presented to or considered by the Examiner during prosecution. While counterpart documents may have been cited, Petitioner asserted there was no indication the Examiner considered the combination or the specific arguments for obviousness presented in the petition.

6. Relief Requested

• Petitioner requested the institution of an IPR and the cancellation of claims 1-20 of the ’873 patent as unpatentable under 35 U.S.C. §103.