Absorbent Article Image Analysis for Individual Biological Prediction

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing methods for managing biological waste and health issues related to absorbent articles, such as diapers, are time-consuming, error-prone, and can lead to health and safety concerns due to improper sizing and fit, with existing systems lacking the ability to accurately analyze individual-specific images for biological features like BM and urination.

Innovation Solution

A biological digital imaging system utilizing artificial intelligence-based learning models to analyze pixel data from digital images of absorbent articles, generating individual-specific prediction values for biological features like stool and urine consistency and amount, providing digital coaching to caregivers.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of information

If manual instruction or manual sizing charts are used to manage biological waste and health issues, then caregivers can access guidance information, but the process becomes time-consuming and error-prone

Engineering Contradiction:

Improveaccuracy of biological feature analysisVSAvoidtime required for manual assessment

Core Design Contradiction:

Loss of informationVSLoss of time

Solution Approach 1:

The patent replaces manual visual inspection and empirical judgment with an automated digital imaging system that uses machine learning models to analyze pixel data. The system automatically detects biological features (stool, urine, skin conditions) in images of absorbent articles, eliminating the need for caregivers to manually assess biological waste characteristics and reducing errors associated with human judgment.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system enables self-service by allowing caregivers to capture images with their mobile devices and receive automated analysis results without requiring professional medical intervention. The machine learning model processes images locally or in the cloud, providing immediate feedback on biological features, diaper fit, and potential health issues, empowering caregivers to make informed decisions independently.

Inventive Principle:

Principle #25Self-service

2Adaptability or versatility

If empirical experimentation with various products or techniques is attempted, then caregivers may find suitable solutions, but health or safety of the infant may be impacted due to trial and error

Engineering Contradiction:

Improveindividual-specific biological predictionVSAvoidhealth and safety outcomes

Core Design Contradiction:

Adaptability or versatilityVSReliability

Solution Approach 1:

The system implements continuous feedback by analyzing images over time and comparing biological feature measurements against individual-specific baselines. The machine learning model tracks changes in stool consistency, urine output, and skin condition, providing caregivers with feedback on whether the current diaper product or care routine is appropriate for their infant's specific needs, eliminating the need for unsafe empirical experimentation.

Inventive Principle:

Principle #23Feedback

Solution Approach 2:

The system performs preliminary analysis by detecting and documenting biological features before health issues develop. By continuously monitoring stool characteristics, urine patterns, and skin condition, the system can identify early signs of problems such as dehydration, allergies, or diaper rash, allowing caregivers to take preventive action rather than reacting to established health issues through trial and error.

Inventive Principle:

Principle #10Preliminary action

3Measurement precision

If digital imaging analysis is implemented to detect biological features, then accuracy in predicting consistency and amount is improved, but system complexity increases

Engineering Contradiction:

Improveprediction accuracy of biological featuresVSAvoidcomplexity of image processing system

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent introduces an intermediary layer consisting of pre-trained machine learning models that specialize in specific biological feature detection tasks. Rather than building a single complex system, the solution uses multiple specialized models (for stool detection, urine detection, skin analysis) that can be independently trained and optimized. These models act as intermediaries between the raw image data and the final predictions, simplifying the overall system architecture while maintaining high accuracy.

Inventive Principle:

Principle #24Intermediary (Mediator)

Data Source

PatentUS12608894B2Digital imaging analysis of biological features detected in physical mediums

Publication Date: 2026.04.21 PROCTER & GAMBLE CO

AI summary

Biological digital imaging systems and methods are disclosed for analyzing pixel data of one or more digital images depicting absorbent articles or portions thereof. A digital image comprising pixel data is obtained depicting an absorbent article or a portion of an absorbent article. An imaging application (app) analyzes the digital image to detect a biological feature depicted within the pixel data of the digital image of the absorbent article or the portion of the absorbent article. The imaging app generates an individual-specific biological prediction value corresponding to at least one of: (a) the absorbent article; (b) the portion of the absorbent article; or (c) an individual associated with the absorbent article or portion of the absorbent article. The individual-specific biological prediction value is based on the biological feature depicted within the pixel data of the digital image of the absorbent article or the portion of the absorbent article.