Align Technology Inc v. Dental Monitoring SAS

1. Case Identification

• Case #: IPR2024-00052
• Patent #: 11,049,248
• Filed: October 16, 2023
• Petitioner(s): Align Technology, Inc.
• Patent Owner(s): Dental Monitoring SAS
• Challenged Claims: 1-2, 4, 7-8, and 13-16

2. Patent Overview

• Title: Method for Analyzing an Image of a Dental Arch
• Brief Description: The ’248 patent discloses a method for assessing the fit of an orthodontic aligner. The method involves acquiring at least one image of a patient wearing the aligner and using a trained deep learning device to analyze the image to determine a value related to the separation between a tooth and the aligner.

3. Grounds for Unpatentability

Ground 1: Claims 1-2, 4, 7-8, and 13-16 are obvious over Kim, Invisalign, Salah, and Maninis.

• Prior Art Relied Upon: Kim (Application # 2017/0281313), Invisalign (2003 Clinical Monitoring Guide), Salah (WO 2016066651), and Maninis (arXiv.org, Apr. 2017).
• Core Argument for this Ground:
  • Prior Art Mapping: Petitioner argued that the prior art collectively discloses every limitation of the challenged claims. Kim taught a method for remote orthodontic services where a patient uses a cellphone to capture and transmit images or videos of a currently worn aligner for an orthodontist to assess. Invisalign provided the clinical basis for this assessment, teaching that aligner fit can be determined by visually inspecting a photograph for a gap of a specific threshold (e.g., >1 mm) between the tooth and the aligner. Salah disclosed using deep learning for contour detection on dental images acquired by a cellphone to monitor orthodontic treatment. Finally, Maninis provided a specific, state-of-the-art deep learning architecture (a Convolutional Neural Network) trained on a large learning base for performing highly accurate, broadly applicable image classification and contour detection. Petitioner asserted that applying Maninis’s advanced contour detection to automate the gap measurement taught by Invisalign within the remote monitoring framework of Kim would render the claimed method obvious.
  • Motivation to Combine: A person of ordinary skill in the art (POSITA) would combine Kim and Invisalign to apply a known, effective method for assessing aligner fit (Invisalign’s gap analysis) to a system for remote monitoring (Kim’s cellphone imaging framework), thereby enhancing accuracy, reducing costs, and facilitating remote assessment. A POSITA would then be motivated to incorporate the deep learning techniques of Salah and Maninis to automate and improve the contour detection and gap measurement process. Maninis taught a flexible and high-performing deep learning architecture that was a logical and advantageous tool for implementing the automated analysis generally suggested by Salah for the specific orthodontic application taught by the Kim/Invisalign combination.
  • Expectation of Success: A POSITA would have had a reasonable expectation of success. The combination involved applying a well-understood and rapidly advancing technology (deep learning for object and contour detection) to a known clinical need (assessing aligner fit from an image) to achieve the predictable result of automated gap measurement. Each reference provided a complementary technique, and their integration was straightforward.

4. Key Claim Construction Positions

• tooth attribute: Petitioner argued this term should be construed broadly to include any attribute related to a tooth that may bear a relation to its separation from an aligner, such as the position of the tooth’s edge, not just a direct measurement of the separation itself.
• image attribute: Petitioner asserted this term encompasses any attribute of the image as a whole that relates to the separation between one or more teeth and the aligner, including the presence or absence of a gap, an average separation value, or an overall fit score.
• image extracted from a film: Petitioner proposed that, in the context of the claims and specification, this phrase should be interpreted to mean an image extracted from a video recording.

5. Arguments Regarding Discretionary Denial

• Petitioner argued that discretionary denial under Fintiv is not warranted. It asserted that the parallel district court proceeding is in its early stages, with no trial date scheduled, no claim construction completed, and no expert reports exchanged.
• Petitioner contended that the expected Final Written Decision (FWD) in this IPR would likely issue before a potential trial in the district court, weighing in favor of institution.
• It was also argued that denial under §325(d) is inappropriate because the petition presents new prior art (Kim, Invisalign, Maninis) and arguments that were not before the Examiner during prosecution. While Salah was cited, it was submitted in French without a full English translation, and its teachings on deep learning were not considered.

6. Relief Requested

• Petitioner requests the institution of an inter partes review and the cancellation of claims 1-2, 4, 7-8, and 13-16 of the ’248 patent as unpatentable.