Insulin Dosing Function System Using Reinforcement Learning

ChangeBridge: Patent Apps - Medical Devices (A61M)

Published September 22nd, 2025

Detected March 26th, 2026

Summary

The USPTO has published a patent application (US20260083909A1) detailing a system and method for calculating an insulin dosing function using reinforcement learning. The application describes a process that utilizes self-attention and a State-Action-Reward-Next State sequence to optimize insulin dosage decisions in automated medical systems.

View original document View source feed page

What changed

This document is a patent application (US20260083909A1) filed with the USPTO, describing a novel system for calculating insulin dosing functions. The core innovation lies in the application of reinforcement learning with self-attention mechanisms to an automated medical system. The system uses a State-Action-Reward-Next State (SARS) sequence, where the state includes continuous glucose monitoring readings, insulin doses, meal information, and activity levels. The agent learns to determine optimal insulin doses by receiving rewards based on resulting glucose levels and updating its neural network weights through iterative learning.

This patent application does not impose immediate regulatory obligations. However, it signals potential future technological advancements in diabetes management and automated insulin delivery systems. Companies involved in developing medical devices, particularly those related to diabetes care and AI-driven health solutions, should be aware of this patent filing as it may impact intellectual property landscapes and future product development strategies in this sector.

Source document (simplified)

← USPTO Patent Applications

SYSTEM AND METHOD FOR CALCULATING AN INSULIN DOSING FUNCTION

Application US20260083909A1 Kind: A1 Mar 26, 2026

Inventors

Anas El Fathi, Marc D. Breton, Elliott C. Pryor, Ali Tavasoli, Heman Shakeri

Abstract

A reinforcement learning process with self attention is used for insulin dosing decisions in an automated medical system. The State-Action-Reward-Next State (SARS) sequence is used. The state represents the current condition, including recent continuous glucose monitoring readings, insulin doses, meal information, and potentially other relevant factors like time of day or physical activity levels. Based on this state, the agent takes an action by deciding on an insulin dose. It then receives a reward, a numerical value quantifying the quality of the action, based on resulting glucose levels and their proximity to the target range. This leads to a new state, and the process repeats. Through this iterative process, the algorithm updates the neural network weights, allowing the agent to learn which actions lead to better outcomes in different states.