PTAB

IPR2022-00227

accessiBe Ltd v. AudioEye Inc

Log In
Key Events
Petition
petition

1. Case Identification

2. Patent Overview

  • Title: Modular Systems and Methods for Selectively Enabling Cloud-Based Assistive Technologies
  • Brief Description: The ’173 patent discloses methods for improving website accessibility for users with disabilities. The technology involves programmatically evaluating a webpage’s Document Object Model (DOM) or HTML code to detect accessibility compliance issues, such as page elements lacking descriptive text, and then automatically assigning appropriate descriptive attributes to those elements for use by assistive technologies like screen readers.

3. Grounds for Unpatentability

Ground 1: Obviousness over Springer and Nguyen - Claims 1, 2, 4, 9-13 are obvious over Springer in view of Nguyen.

  • Prior Art Relied Upon: Springer (Application # 2004/0148568) and Nguyen (a 2008 article titled “Learning to Extract Form Labels”).
  • Core Argument for this Ground:
    • Prior Art Mapping: Petitioner argued Springer teaches a system for automatically remediating web accessibility issues. It uses a set of checker tools (e.g., a "FormChecker") to find compliance problems like untagged form fields and corresponding fixer tools (e.g., a "FormFixer") that use heuristic rules to identify and apply a descriptive label from nearby text. Nguyen discloses using a superior machine learning (a type of artificial intelligence) approach that analyzes various contextual cues (e.g., proximity, layout patterns, element type) to more accurately extract labels for untagged form elements.
    • Motivation to Combine: A Person of Ordinary Skill in the Art (POSITA) would combine these references to improve the accuracy of Springer's heuristic-based remediation tools. Nguyen expressly presents its machine-learning method as superior to prior heuristic-based systems. A POSITA would therefore be motivated to replace Springer's heuristic "FormFixer" logic with Nguyen's more robust machine learning algorithm to better identify and assign labels to untagged form fields.
    • Expectation of Success: A POSITA would expect this combination to succeed with minimal modification, as it involves applying a known, improved technique (machine learning for label extraction) to a known problem within an existing accessibility remediation framework.

Ground 2: Obviousness over Springer, Nguyen, and Bigham - Claims 7 and 8 are obvious over Springer and Nguyen in view of Bigham.

  • Prior Art Relied Upon: Springer (Application # 2004/0148568), Nguyen (2008 article), and Bigham (a 2007 article titled "Increasing Web Accessibility by Automatically Judging Alternative Text Quality").
  • Core Argument for this Ground:
    • Prior Art Mapping: This ground builds on Ground 1 and specifically addresses claims for untagged graphics and photographs. Springer discloses tools like "ImageChecker" and "ImageFixer" for remediating images that lack alternative ("alt") text. Bigham teaches a machine learning classifier that automatically generates and judges the quality of alt text for images by considering the context surrounding the image on the webpage.
    • Motivation to Combine: The motivation is analogous to Ground 1. Just as Nguyen improves label extraction for forms, Bigham provides a technologically superior, machine-learning-based method for handling alt text for images. A POSITA would be motivated to integrate Bigham's trained classifier into Springer’s system to improve the quality and accuracy of the alt text generated by the "ImageFixer," thereby creating a more effective accessibility solution for graphical elements.

Ground 3: Obviousness over Springer, Nguyen, and Hendry - Claims 3, 5, 16-19, 24-28 are obvious over Springer and Nguyen in view of Hendry.

  • Prior Art Relied Upon: Springer (Application # 2004/0148568), Nguyen (2008 article), and Hendry (Application # 2013/0104029).
  • Core Argument for this Ground:
    • Prior Art Mapping: This ground addresses claims directed to a distributed computing environment (e.g., involving a remote server) and the use of JavaScript. The base combination of Springer and Nguyen provides the AI-driven accessibility remediation system. Hendry teaches a system for automated accessibility remediation that operates in a distributed client-server architecture, where tasks can be split between a local user computer and a remote server to leverage the server's greater processing power and storage. Hendry also explicitly discloses using JavaScript as the remediation code.
    • Motivation to Combine: A POSITA would combine Hendry’s architecture with the Springer/Nguyen system for well-understood benefits. Artificial intelligence and machine learning processes can be computationally intensive. A POSITA would be motivated to offload these tasks to a remote server, as taught by Hendry, to improve performance on the user's local machine. This was a common and predictable design choice for implementing computationally heavy web applications.
  • Additional Grounds: Petitioner asserted additional obviousness challenges based on Lehota (Application # 2010/0205523) as a primary reference in combination with Nguyen, Bigham, and Hendry, but relied on similar technological substitution and design modification theories.

4. Arguments Regarding Discretionary Denial

  • Petitioner argued that discretionary denial under §314(a) based on the Fintiv factors would be inappropriate. The parallel district court case was in a very early stage, with fact discovery having only recently opened and no substantive investment by the court in the ’173 patent.
  • Petitioner contended that the district court's trial date, scheduled approximately 10 months from the petition's filing, was tentative and statistically likely to be rescheduled to a date after the Board’s one-year deadline for a Final Written Decision (FWD).
  • The petition asserted it raised materially different prior art combinations and invalidity arguments than those presented in preliminary filings in the district court. Finally, Petitioner argued that the strong merits of its unpatentability challenges weighed heavily in favor of institution.

5. Relief Requested

  • Petitioner requests institution of an inter partes review (IPR) and cancellation of claims 1-5, 7-13, 16-20, and 22-28 of the ’173 patent as unpatentable.