Automated Insulin Dosing via Gesture-Based Meal Detection

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current systems for managing insulin dosing in Type 1 diabetes patients are inefficient due to delays in glucose readings and insulin diffusion, requiring manual input for meal announcements, which can lead to poor glycemic control and adherence issues.

Innovation Solution

An automated medication dispensing system using sensors to detect movement and gestures, processing data to determine food intake events and adjust medication administration, including reminders and communication with insulin management systems.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If manual input is used for meal announcements, then the system can respond to eating events, but the patient burden increases and compliance decreases

Engineering Contradiction:

Improveease of meal announcementVSAvoidautomation of meal detection

Core Design Contradiction:

Ease of operationVSExtent of automation

Solution Approach 1:

The system automatically detects eating events using sensors (accelerometer, gyroscope, camera) without requiring patient input. The meal detection system monitors gestures, facial expressions, and utensil movements to autonomously identify when a patient is eating, eliminating the need for manual meal announcements while maintaining accurate detection

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The patent replaces manual mechanical input (patient pressing buttons or speaking) with automated sensor-based detection. Multiple sensors including accelerometers, gyroscopes, and cameras work together to detect eating events automatically, substituting the mechanical action of manual announcement with electronic sensing and processing

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Ease of operation

If automated meal detection is implemented, then patient burden is reduced, but system complexity increases

Engineering Contradiction:

Improveease of meal announcementVSAvoidcomplexity of detection system

Core Design Contradiction:

Ease of operationVSDevice complexity

Solution Approach 1:

The meal detection system is divided into separate functional modules: gesture detection module, facial expression analysis module, utensil detection module, and confidence threshold module. Each module processes specific sensor data independently before integrating results, making the complex system manageable and maintainable while achieving accurate meal detection

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The sensor system serves multiple functions simultaneously - the accelerometer detects both gestures and utensil movement, the gyroscope tracks head movements and eating motions, and the camera captures both facial expressions and food consumption. This multi-functionality reduces the need for separate dedicated sensors for each detection task

Inventive Principle:

Principle #6Universality (Multi-functionality)

3Reliability

If real-time meal detection is achieved, then glycemic control improves, but processing speed requirements increase

Engineering Contradiction:

Improveaccuracy of glycemic controlVSAvoidspeed of event detection

Core Design Contradiction:

ReliabilityVSSpeed

Solution Approach 1:

The system performs preliminary actions by continuously monitoring sensor data in advance of actual meal consumption. Gestures such as bringing utensils to the mouth or picking up food containers are detected before the actual eating begins, allowing the system to prepare insulin dosing recommendations in advance and reduce processing delays

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The sensor system operates continuously, constantly analyzing gestures, facial expressions, and utensil movements without interruption. This continuous monitoring ensures that eating events are detected immediately when they occur, maintaining real-time readiness for insulin dosing adjustments and eliminating detection delays

Inventive Principle:

Principle #20Continuity of useful action

Data Source

PatentUS11367517B2Gesture-based detection of a physical behavior event based on gesture sensor data and supplemental information from at least one external source

Publication Date: 2022.06.21 MEDTRONIC MINIMED INC

AI summary

An automated medication dosing and dispensing system includes: gesture sensors to detect physical movement of a user; computer-readable storage media with program code instructions; and at least one processor. The instructions are configurable to cause the at least one processor to perform a method that involves: detecting, based on analysis of gesture sensor readings obtained from the gesture sensors, an occurrence of a gesture-based physical behavior event; calculating an initial confidence level for the detected occurrence of the gesture-based physical behavior event; deriving a final confidence level, for the detected occurrence of the gesture-based physical behavior event, from the initial confidence level and from external information obtained from at least one source of data that is distinct from the gesture sensors; and adjusting medication dosage and/or medication dispensing parameters in response to the detected occurrence of the gesture-based physical behavior event and the final confidence level.