AI Visualize asserted U.S. Patent Nos 8,701,167; 9,106,609; 9,438,667; and 10,930,397 against Nuance Communications and Mach 7 in the District Court of Delaware.  U.S. Patent No. 9,106,609 ('609 patent) was determined as the representative patent.

The asserted patents relate to visualization of medical scans.  Medical imaging systems such as MRI scans create a collection of two-dimensional cross-section images. '609 patent, 1:27-32. Images are stored at a central server as 3D collection of data referred to as a "volume visualization dataset" (VVD). '609 patent, 1:27-32. Viewing large VVDs at the client side requires a user's computer to have high computing capacity. The claimed invention is directed to a method and system of a common and centralized infrastructure, for receiving, storing, processing and viewing large medical scans via a low-bandwidth web portal.  '609 patent, 1:58-62.  Particularly, the claimed invention recites, inter alia  “creating the requested frames of the requested views from the volume visualization dataset in the central storage medium.”

Nuance and Mach7 moved to dismiss under FRCP 12(b)(6) for failure to state a claim alleging that the asserted patent claims were directed to patent-ineligible subject matter and therefore, invalid under 35 U.S.C. § 101. The district court granted the motion finding that the asserted patent claims were directed to an abstract idea and failed to provide an inventive step that transformed the abstract idea into a patent eligible invention.

The Federal Circuit upheld the District Court’s holding finding that the asserted claims recite patent ineligible subject matter.  The Federal Circuit generalized the claims into the following functional steps: (i) storing data on a server; (ii) accepting a user request to view a portion of that data; (iii) checking for the location of all data needed for the virtual view; (iv) creating image frames from non-locally-stored frames; (v) compiling all image frames; and (vi) sequentially displaying the image frames to the user.  The Federal Circuit concluded that the asserted claims are directed to converting data and using computers to collect, manipulate, and display the data.  Particularly, the Federal Circuit noted that the steps of obtaining, manipulating, and displaying data, particularly when claimed at a high level of generality are abstract concepts (citing Elec. Power Grp., LLC v. Alstom S.A., 830 F.3d 1350, 1353-54 (Fed. Cir. 2016)) and converting information from one format to another is an abstract idea (citing Hawk Tech. Sys., LLC v. Castle Retail, LLC, 60 F.4th 1349, 1357 (Fed. Cir. 2023)).

AI Visualize argued that (i) the claims are not directed to an abstract idea because the claims are directed to creation of "on the fly" virtual views at a client computer, and (ii) the '609 patent contains multiple disclosures supporting the view that the "creating" step provides a technical solution to a technical problem.  The Federal Circuit rejected both of these arguments noting that the claimed "creating" operation is achieved by manipulation of a portion of an existing VVD.  The Federal circuit further noted that the claims do not specify the manner that meaningfully supports a technical solution to a technical problem in the prior art.

AI Visualize further argued that  the creation of virtual views transforms the claims into patent-eligible subject matter since the creation of virtual views "on demand" or in "real-time" in response to a user request transforms the claims into something "significantly more" than the abstract idea.  The Federal Circuit also rejected this argument noting that the claimed step of creating a virtual view is itself the abstract idea rather than "significantly more" than the abstract idea.

Notably, the lack of disclosure in the specification of the ’609 Patent likely played a significant role in the outcome of the case.  In this regard, the disclosure in the ’609 Patent regarding how to create the virtual views was too generic.  For example, while the ’609 Patent at 14:4-10 generally describes using machine learning algorithms and neural networks for processing volume visualization datasets, no specific technical embodiments were disclosed regarding how to create the virtual views.  As a result, the asserted claims recited features with a high level of functionality, which ultimately lead to the Federal Circuit concluding that the asserted claims were directed to patent ineligible subject matter.