Parameter-Invariant Meal Detection Algorithm for Artificial Pancreas

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current artificial pancreas systems face challenges in accurately and timely detecting meal events due to their reliance on patient-specific physiological parameters, leading to high false positives and detection delays, which can result in life-threatening hypoglycemia or postprandial hyperglycemia.

Innovation Solution

A physiology parameter-invariant meal detection algorithm that uses a minimal glucose physiological model and null space projection to detect meal events, independent of individual physiological parameters, achieving a near constant false alarm rate across the patient population.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If patient-specific physiological parameters are used for meal detection, then detection accuracy for individual patients may improve, but false positives increase and detection delays occur

Engineering Contradiction:

Improvemeal detection accuracyVSAvoidfalse alarm rate

Core Design Contradiction:

Measurement precisionVSReliability

Solution Approach 1:

The patent transforms the meal detection problem from using patient-specific physiological parameters to using parameter-invariant features. By changing the detection parameters from individualized physiological constants to universal meal-induced glucose response patterns, the system achieves consistent performance across patients without requiring patient-specific calibration, thereby reducing false positives while maintaining detection accuracy

Inventive Principle:

Principle #35Parameter changes

Solution Approach 2:

The patent segments the glucose response into distinct phases: baseline glucose level, meal-induced glucose rise, and post-meal glucose trajectory. By analyzing these segmented phases separately and identifying characteristic patterns in each, the system can detect meals accurately without being confounded by patient-specific physiological variations, thus improving reliability

Inventive Principle:

Principle #1Segmentation

2Measurement precision

If patient-specific physiological parameters are used for meal detection, then individualized detection may be achieved, but detection delays occur

Engineering Contradiction:

Improvemeal detection accuracyVSAvoiddetection delay

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The patent establishes baseline glucose levels and typical meal response patterns before actual meal detection is needed. By pre-characterizing the glucose system's response to meals using parameter-invariant features, the system is prepared to rapidly detect meals without requiring time-consuming individual parameter identification during the detection process, thereby reducing detection delays

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent skips the traditional multi-step process of identifying patient-specific parameters by directly applying parameter-invariant meal detection algorithms. This approach rushes through the detection process by using universal meal response patterns that can be immediately applied without individual calibration, significantly reducing the time from meal ingestion to detection

Inventive Principle:

Principle #21Skipping (Rushing through)

3Reliability

If physiology parameter-invariant algorithm is used, then consistent performance across patients is achieved, but individual physiological variations may be overlooked

Engineering Contradiction:

Improveperformance consistencyVSAvoidphysiological adaptability

Core Design Contradiction:

ReliabilityVSAdaptability or versatility

Solution Approach 1:

The patent applies universality by developing a meal detection algorithm that works across all patients without requiring individual customization. The parameter-invariant features capture universal aspects of meal-induced glucose responses that are common to all patients, enabling the same algorithm to function effectively for diverse patient populations while maintaining consistent performance

Inventive Principle:

Principle #6Universality (Multi-functionality)

Solution Approach 2:

The patent introduces parameter-invariant meal detection features as an intermediary between raw glucose data and meal detection decisions. These intermediary features capture the essential meal-induced patterns while filtering out patient-specific physiological variations, allowing the system to achieve both consistency across patients and adaptability to individual meal responses

Inventive Principle:

Principle #24Intermediary (Mediator)

Data Source

PatentUS10792423B2Methods, systems, and computer readable media for physiology parameter-invariant meal detection

Publication Date: 2020.10.06 THE TRUSTEES OF THE UNIV OF PENNSYLVANIA

AI summary

Methods, systems, and computer readable media for physiology parameter-invariant meal detection are disclosed. According to one system, the system includes at least one processor and a meal detection module implemented using the at least one processor. The meal detection module is configured to receive insulin intake information and blood glucose level information for a user, to detect a meal event using a physiology parameter-invariant meal detection algorithm, and after detecting the meal event, to perform at least one control action associated with insulin management.