Healthcare Management System Patent Application Filed

ChangeBridge: Patent Apps - Health Informatics (G16H)

Published September 24th, 2025

Detected March 26th, 2026

Summary

The USPTO has published a patent application (US20260088145A1) for a healthcare management system. The application, filed on September 24, 2025, describes a computer system for analyzing medical reports using artificial intelligence and vectorized data embeddings to identify injury categories and severity.

View original document View source feed page

What changed

The United States Patent and Trademark Office (USPTO) has published a patent application (US20260088145A1) for a healthcare management system. The application details a computer system designed to process user-supplied text and medical report data, utilizing AI to extract injury descriptions, vectorize them, and compare them against a database of predefined labels and severity codes to structure the results.

This publication represents a new patent application and does not impose any immediate regulatory obligations or compliance requirements on healthcare providers or technology companies. It signifies a development in health informatics technology, potentially impacting future product development and data analysis methodologies within the healthcare sector. Compliance officers should note this as a technological advancement rather than an actionable regulatory change.

Source document (simplified)

← USPTO Patent Applications

HEALTHCARE MANAGEMENT SYSTEM

Application US20260088145A1 Kind: A1 Mar 26, 2026

Inventors

Marc E. Jackson, Jackson M. Neeley

Abstract

A computer system for structured analysis of medical reports includes one or more processors and computer-readable media storing instructions that configure the system to process user-supplied text and medical report data. Written text, including linguistic variants and descriptive phrases associated with injury categories, is received and vectorized into a first embedding vector. The text and embedding vector are stored in a hybrid database comprising predefined labels corresponding to injury categories and severity codes, free-form text fields storing linguistic variants and descriptive phrases, and stored vector embeddings generated for the free-form text fields. A medical report is received via an artificial intelligence engine trained to extract a description of a patient injury, and the description is vectorized into a second embedding vector. The second embedding vector is compared against the stored vector embeddings within the hybrid database, one or more injury categories are selected based on the comparison, and a structured result including at least a predefined label is output.