Wearable Actigraphy Anomaly Detection for Depression Relapse Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Major Depressive Disorder (MDD) often goes undetected until symptoms worsen, leading to delayed treatment and increased risk of self-harm or suicide, due to the reactive nature of current clinical approaches that fail to monitor early changes in patients' symptomatology between physician visits.
Innovation Solution
A computer-implemented method and system using wearable devices to collect actigraphy data and train anomaly detectors, which analyze movement patterns, sleep, and circadian rhythms to predict the onset of depression through machine-learning algorithms and self-report tests, enabling early identification of relapse or recurrence.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If clinicians use a reactive approach by observing patients only during clinical visits, then device complexity is reduced, but detection precision of early depression changes deteriorates
Solution Approach 1:
The patent replaces manual clinical observation with automated machine learning algorithms that analyze actigraphy data from wearable devices. The anomaly detector automatically identifies deviations from baseline movement patterns, substituting the mechanical process of clinical visits with computational analysis that operates continuously without human intervention.
Solution Approach 2:
The system enables self-monitoring where the wearable device continuously collects actigraphy data and the machine learning model automatically detects anomalies without requiring clinician intervention. The system serves itself by maintaining baseline profiles and autonomously identifying when patients deviate from normal patterns, enabling continuous monitoring without increasing operational complexity.
2Measurement precision
If clinicians increase monitoring frequency between visits, then detection precision improves, but loss of time increases
Solution Approach 1:
The system implements continuous monitoring of actigraphy data between clinical visits, eliminating gaps in observation. The wearable device continuously tracks movement patterns, and the machine learning model continuously analyzes this data stream, ensuring that depression relapse is detected as soon as it begins rather than waiting for the next scheduled visit.
Solution Approach 2:
The system performs preliminary detection of depression relapse by identifying subtle changes in movement patterns before they manifest as overt symptoms. The anomaly detector flags early deviations from baseline behavior, allowing clinicians to intervene before the patient's condition significantly worsens, thereby reducing the time lost to undetected deterioration.
3Measurement precision
If clinicians implement continuous monitoring, then detection precision improves, but device complexity increases
Solution Approach 1:
The patent extracts only the essential monitoring function from complex clinical assessment tools. Instead of implementing comprehensive continuous clinical evaluation, the system isolates actigraphy data collection and anomaly detection, removing unnecessary complexity while maintaining effective monitoring capability. The wearable device focuses solely on movement tracking, and the algorithm focuses solely on detecting deviations from baseline patterns.
4Reliability
If clinicians use simple observation methods, then device complexity is reduced, but reliability of depression detection deteriorates
Solution Approach 1:
The system implements feedback loops where the machine learning model continuously compares current actigraphy data against established baseline profiles. When anomalies are detected, the system can trigger alerts to clinicians or prompt patients to complete self-report assessments, creating a feedback mechanism that enhances detection reliability. The model also learns from confirmed cases to improve future detection accuracy.
Data Source
AI summary
A system and computer-implemented method for detecting return of depression of a patient is provided. The system comprises a wearable device configured to detect movement of the patient and configured to generate actigraphy data corresponding to the movement of the patient and a computing device for retrieving actigraphy data from the device. The system and method obtain training data, including training actigraphy data, over a training period and train an anomaly detector using the training data. The system and method subsequently obtain test data from the patient, extract a plurality of features from the test data, and analyze the extracted data using the trained anomaly detector. A self-report test is used to determine whether an anomaly identified by the anomaly detector indicates that the patient is likely to experience return of depression.