Video Actigraphy for Delirium Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current delirium detection methods in ICU patients are inadequate due to under-detection and lack of continuous monitoring, as they rely on infrequent screenings and wrist actigraphy, which misses movements and can irritate patients, leading to delayed diagnosis and increased healthcare costs.
Innovation Solution
A monitoring system using video actigraphy to capture full-body motion data, analyzing image data for delirium-typical motion events, and calculating a delirium score based on duration, intensity, type, location, and occurrence of these events, providing continuous and objective monitoring without additional sensors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If wrist actigraphy is used to measure motor activity, then motoric alterations can be detected, but movements of other body parts are missed leading to under-detection of delirium
Solution Approach 1:
The patent divides the body into multiple monitored segments by placing accelerometers on different body parts (wrist, ankle, trunk). This segmentation allows comprehensive detection of motoric alterations across the entire body, overcoming the limitation of single-point monitoring and enabling accurate detection of delirium-related movements in any body region.
2Productivity
If on-body sensors are attached to patients for continuous monitoring, then delirium can be detected continuously, but the sensors may irritate or confuse patients
Solution Approach 1:
The patent extracts the monitoring function from direct body contact by using video cameras to capture motor activity from a distance. This eliminates the need for worn sensors that cause patient irritation or confusion, while maintaining continuous objective monitoring capabilities through non-contact video analysis of body movements.
3Measurement precision
If screening questionnaires are used to diagnose delirium, then delirium can be detected, but patients are checked only three times a day leading to missed delirious episodes
Solution Approach 1:
The patent implements continuous monitoring by using video cameras and accelerometers to track motor activity 24/7 without interruption. This continuous data collection captures all delirious episodes regardless of frequency, eliminating the gaps inherent in periodic questionnaire-based screening and enabling real-time detection of delirium onset and progression.
4Adaptability or versatility
If video monitoring is used for whole body motion detection, then all body movements can be captured, but the system complexity increases compared to wrist actigraphy
Solution Approach 1:
The patent merges multiple monitoring approaches by integrating video camera data with accelerometer measurements from wearable devices. This combination leverages the comprehensive body coverage of video monitoring while using accelerometers to simplify detection of specific movement patterns, reducing the overall system complexity while maintaining complete body monitoring capability.
Data Source
AI summary
The present invention relates to a monitoring system (1, 2, 3) and a corresponding monitoring method for monitoring a patient and detecting delirium of the patient in an unobtrusive manner without the need of on-body sensors. The proposed monitoring system comprises a monitoring unit (10) for obtaining image data (30) of the patient over time, an image analysis unit (12) for detecting motion events of the patient from the obtained image data (30), an evaluation unit (14) for classifying the detected motion events into delirium-typical motion events and non-delirium-typical motion events, and a delirium determination unit (16) for determining a delirium score (32) from the duration, intensity, type, location and/or occurrence of delirium-typical motion events, said delirium score (32) indicating the likelihood and/or strength of delirium of the patient.