Physical Condition Detection Using Anomaly Scoring
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing health monitoring techniques fail to detect small changes in physical conditions that may lead to anomalies in health, particularly in aging populations, leading to potential severe illness due to overlooked signs.
Innovation Solution
A physical condition detection method that uses a computer to obtain activity data, calculate features, and input them into a model trained on normality or anomaly patterns to generate an anomaly score, which is then graded to indicate the level of physical condition anomaly, enabling early detection of health anomalies.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a monitoring system only notifies when vital information indicates an anomalous value, then false alarms are reduced, but small changes in physical condition that may lead to health anomalies are overlooked
Solution Approach 1:
The system performs preliminary analysis by calculating multiple features from activity data and inputting them into a trained model that has learned normality or anomaly patterns. This preliminary processing enables the system to detect small changes in physical conditions before they develop into significant health anomalies, allowing early intervention while maintaining reliable detection accuracy.
Solution Approach 2:
The system transforms raw activity data into multiple calculated features and processes them through different stages (feature extraction, model input, anomaly score calculation, graded score generation). These parameter transformations enable the system to detect subtle physical condition changes that would be imperceptible in the original data while maintaining detection reliability through the trained model.
2Measurement precision
If health care workers manually monitor all subjects, then small changes in physical condition can be detected, but the labor shortage in medical and caregiving services cannot be addressed
Solution Approach 1:
The monitoring system performs self-service by automatically obtaining activity data, calculating features, inputting them into a trained model, calculating anomaly scores, and generating graded scores without requiring continuous human intervention. This automation enables the system to detect small physical condition changes independently, addressing the labor shortage while maintaining detection precision that would otherwise require manual monitoring.
Solution Approach 2:
The system replaces the mechanical manual monitoring process with an automated computer-based system that obtains activity data, calculates features, processes them through a trained model, and generates anomaly assessments. This substitution eliminates the need for increased human labor while maintaining or improving the detection of small physical condition changes.
3Measurement precision
If a model trained on activity data groups is used to calculate anomaly scores, then detection accuracy improves, but the complexity of the system increases
Solution Approach 1:
The system segments the anomaly detection process into distinct modules: obtaining activity data, calculating multiple features from the data, inputting features into a trained model, calculating anomaly scores from model output, and generating graded scores. This segmentation manages system complexity by organizing functions into separate, manageable components while maintaining high detection accuracy through the trained model.
Solution Approach 2:
The trained model acts as an intermediary between raw activity data and anomaly detection results. The model receives multiple calculated features as input and produces anomaly scores that reflect the degree of anomaly in physical conditions. This intermediary layer simplifies the overall system architecture while enabling accurate anomaly detection through the model's learned patterns of normality and anomaly.
Data Source
AI summary
A physical condition detection method performed by a computer includes: obtaining activity data including a respiratory rate and a heart rate of a subject during a predetermined time period; calculating a plurality of features, based on the activity data obtained; obtaining an anomaly score indicating a degree of an anomaly in a physical condition per the predetermined time period, by inputting the plurality of features calculated into a model that has learned normality or anomaly in an activity data group; calculating a graded score for indicating a physical condition anomaly level of the subject in a graded manner, based on the anomaly score obtained; and outputting the graded score calculated.