Smart Insulin Pen with Image Recognition for Dose Calculation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional insulin dose calculators for diabetes management require manual estimation of carbohydrates, which can be inaccurate and burdensome, leading to non-compliance and potential health risks due to incorrect dosing.

Innovation Solution

A system and method for a smart medicine delivery pen and software application that provides a touchless, fixed-dose titration protocol for estimating insulin doses based on meal type and size, automatically logging doses, and adjusting insulin delivery to maintain target glucose levels, reducing user input and increasing accuracy.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If manual carbohydrate estimation is used in conventional dose calculators, then users can obtain insulin dosing recommendations, but the method is burdensome and inaccurate leading to non-compliance

Engineering Contradiction:
Improvecarbohydrate estimation accuracyVSAvoiduser burden
Core Design Contradiction:
Measurement precisionVSEase of operation

Solution Approach 1:

The patent extracts the carbohydrate estimation task from manual user input and assigns it to image recognition technology. The system captures images of meals and automatically identifies and quantifies carbohydrate content, removing the burden of manual counting while maintaining or improving accuracy through automated visual analysis.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The system enables self-service by allowing the meal image to automatically provide the carbohydrate information needed for dosing calculations. The image recognition algorithm independently analyzes the meal composition and inputs the data into the dose calculator without requiring user intervention or expertise in carbohydrate counting.

Inventive Principle:
Principle #25Self-service

2Ease of operation

If simplified meal estimation methods are used, then user compliance increases, but dosing precision may be reduced

Engineering Contradiction:
Improveuser complianceVSAvoidmeal estimation accuracy
Core Design Contradiction:
Ease of operationVSMeasurement precision

Solution Approach 1:

The patent replaces the mechanical process of manual carbohydrate counting with an optical system (image recognition). This substitution maintains simplicity for the user while achieving higher precision through automated visual analysis of meal composition, eliminating the trade-off between ease of use and accuracy.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system changes the input parameter from manual numerical input to visual image data. By capturing images and analyzing them computationally, the system transforms the interaction mode while maintaining user simplicity, allowing complex analysis to occur automatically in the background without requiring user expertise.

Inventive Principle:
Principle #35Parameter changes

3Measurement precision

If fixed-dose titration with meal type and size estimation is used, then dosing accuracy improves, but system complexity increases

Engineering Contradiction:
Improveinsulin dosing accuracyVSAvoidsystem complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent implements a universal image recognition system that handles multiple meal types and sizes through a single automated process. The same technology platform processes diverse meals by identifying food categories and portions, providing accurate dosing recommendations without requiring separate manual protocols for different meal scenarios.

Inventive Principle:
Principle #6Universality (Multi-functionality)

Solution Approach 2:

The system performs self-service by automatically analyzing meal images to determine type and size, then calculating the appropriate insulin dose. This eliminates the need for users to navigate complex decision trees or input multiple parameters manually, as the system independently completes the entire analysis and calculation process.

Inventive Principle:
Principle #25Self-service

4Reliability

If automated dose logging is implemented, then compliance monitoring improves, but data processing requirements increase

Engineering Contradiction:
Improvedose logging accuracyVSAvoiddata processing load
Core Design Contradiction:
ReliabilityVSUse of energy by moving object

Solution Approach 1:

The patent merges the dose calculation and logging functions into a single automated workflow. The same image recognition system that determines carbohydrate content also captures meal context data, which is then integrated with dosing information for comprehensive logging. This consolidation reduces redundant data processing while improving compliance monitoring.

Inventive Principle:
Principle #5Merging (Combining)

Data Source

PatentUS20220172816A1Methods, systems and devices for a medicament dose calculator
Publication Date: 2022.06.02 MEDTRONIC MINIMED INC
  • US20220172816A1 patent drawing
  • US20220172816A1 patent drawing
  • US20220172816A1 patent drawing

AI summary

Disclosed are systems and techniques providing alternatives to traditional insulin dose calculators that involve little to no input from a patient user of a medicine injection device, including a fixed-dose titration therapy dose calculator and a streamlined dose calculator. In some aspects, a system for administering a medicine using a fixed-dose titration protocol includes an injection pen device in wireless communication with a mobile communication device comprising a software application to determine a recommended one or more fixed-dose sizes of the insulin based on (i) health data, including first glucose level of a patient user of the injection pen device that is measured prior to consumption of a meal and a second glucose level of the patient user that is measured within a predefined time period after consumption of the meal, and (ii) meal data, including a meal type of the meal and a meal size of the meal.