Patient Exit Detection Using Motion Pattern Matching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current automated patient monitoring systems in healthcare facilities face high levels of false positives and false negatives in detecting patient bed exiting events due to a 'one size fits all' approach, failing to account for individual patient movements and habits, leading to inadequate supervision and increased risk of falls.
Innovation Solution
A computer system that accesses movement data from sensors to generate a motion capture pattern summary, compares it to a library of movement pattern data sets, and initiates remedial actions such as lowering the support platform or raising bedrails when a specific pattern indicative of exiting is detected, tailored to each patient's unique behavior.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a one size fits all approach is used to detect patient movements, then the monitoring system is simple to implement, but the detection accuracy is low with high false positives and false negatives
Solution Approach 1:
The system changes the parameter of movement pattern recognition from generic thresholds to patient-specific patterns. Each patient's movement characteristics (speed, amplitude, sequence) are captured and stored as unique parameters, allowing the system to distinguish between normal movements and exit attempts with high accuracy while maintaining manageable complexity through automated pattern learning.
Solution Approach 2:
The system performs preliminary action by capturing and storing each patient's movement patterns during normal activities before actual exit attempts occur. This baseline data is used to train the recognition algorithm, enabling the system to accurately distinguish between routine movements and genuine exit attempts, thereby improving detection accuracy without requiring complex real-time analysis.
2Reliability
If continuous direct supervision is provided for every patient, then patient safety is maximized, but the cost and logistical burden becomes prohibitive
Solution Approach 1:
The monitoring system enables self-service by allowing patients to be monitored independently through automated sensors and pattern recognition. The system autonomously detects exit attempts and alerts caregivers only when necessary, eliminating the need for continuous human supervision while maintaining high patient safety. This significantly improves caregiver efficiency by reducing unnecessary interventions and allowing them to focus on patients who actually need assistance.
3Measurement precision
If automated monitoring systems use generic movement detection, then the system is easy to implement, but it produces high levels of false positives causing caregivers to ignore true positives
Solution Approach 1:
The system performs preliminary action by capturing and storing each patient's movement patterns during normal activities before actual exit attempts occur. This baseline data is used to train the recognition algorithm, enabling the system to accurately distinguish between routine movements and genuine exit attempts, thereby improving detection accuracy without requiring complex real-time analysis.
Solution Approach 2:
The system implements feedback by continuously comparing detected movements against stored patient-specific patterns and adjusting its recognition thresholds based on accumulated data. This learning mechanism reduces false positives over time while maintaining ease of operation, as the system automatically refines its detection accuracy without requiring manual recalibration or complex user intervention.
Data Source
AI summary
The present invention relates to systems and methods for monitoring patient support exiting and initiating a response. Movement data is accessed from sensors (e.g., cameras) that are monitoring a patient resting on a support platform. A motion capture pattern summary is generated from the accessed movement data. The motion capture pattern summary is compared to one or more movement pattern data sets in a library of movement pattern data sets. It is determined that the motion capture pattern summary is sufficiently similar to one of the one or more movement pattern data sets in the library of movement pattern data sets. From the determined similarity it is determined that the patient is attempting to exit the support platform. Remedial measures are initiated to prevent the detected platform support exiting attempt.