Cross section views of wounds

Inventors

ADIRI, Yonatan • Zohar, Ron • Omer, Ido • Bubis, Nathaniel

Assignees

HealthyIo Ltd

Abstract

A non-transitory computer readable medium storing data and computer implementable instructions that, when executed by at least one processor, cause the at least one processor to perform operations for generating cross section views of a wound, the operations including receiving 3D information of a wound based on information captured using an image sensor associated with an image plane substantially parallel to the wound; generating a cross section view of the wound by analyzing the 3D information; and providing data configured to cause a presentation of the generated cross section view of the wound.

Core Innovation

The invention generates cross section views of a wound from 3D information captured using an image sensor associated with an image plane substantially parallel to the wound. The generated output includes a plurality of parallel cross section views of the wound, based on analyzing the 3D information, and the cross section views are provided as data configured to cause a presentation of the wound views.

The disclosure includes using 3D information to produce wound-related visual representations, including cross section views and visual indications tied to wound depth and tissue-related information. It also describes computing and presenting tissue and depth indicators and wound condition overlays over time, including interpolation or extrapolation of wound condition indicators, conditioned on a treatment cycle and treatment phase.

The disclosure further describes selecting and/or generating synthetic visual content associated with a wound over a visual time series using simulated viewpoint properties. This includes generating simulated wound images at a specified viewpoint and point in time when a direct image does not exist, using video inpainting or a generative model, including a transformer-based generative model.

A simulated virtual camera trajectory is used to rearrange frames and select or create synthetic frames, and convolution-based computations are used to generate overlay indication data. The disclosure also includes depth determination and skin-position estimation using convolution-based depth estimation and inpainting, including semantic segmentation to obtain tissue types and wound depth.

Claims Coverage

The document provides three independent claim sets: a non-transitory computer readable medium, a system, and a computer-implemented method. Across these claims, there are two inventive features: receiving wound 3D information from an image sensor with an image plane substantially parallel to the wound, and generating and presenting a wound cross section view comprising a plurality of parallel cross section views.

Each independent claim is directed to generating a wound cross section view comprising a plurality of parallel cross section views from 3D wound information captured with an image plane substantially parallel to the wound, and then providing data configured to cause a presentation of the generated cross section view(s).

Stated Advantages

Documented Applications