Environmental Sensor Fusion for Care Emergency Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing personal emergency response systems (PERS) struggle with false positives and negatives in detecting emergencies, leading to unnecessary responses or missed alerts, especially in environments where multiple sensors are involved.
Innovation Solution
A system comprising care analytics management processors (CAMP) and environmental sensors that dynamically configure sensor thresholds and data storage based on detected edge conditions, using elastic repositories and edge devices to verify sensor data, reducing false positives and negatives through integrated sensor communication and pattern recognition.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If multiple environmental sensors are deployed to monitor the person under care, then the reliability of emergency detection is improved, but the number of false positives increases due to conflicting sensor data
Solution Approach 1:
The system implements feedback loops where sensor data is continuously monitored and evaluated against learned normal patterns. When edge conditions are detected, the system adjusts sensor thresholds dynamically based on feedback from the elastic repository, which stores historical sensor data and patterns. This feedback mechanism allows the system to distinguish between actual emergencies and false positive conditions, reducing false alarms while maintaining high detection reliability.
Solution Approach 2:
The patent dynamically changes sensor detection parameters and thresholds based on the detected state. During quiescent periods, sensors operate with standard thresholds, but when edge conditions are identified, the system adjusts thresholds and sensitivity parameters to account for the specific condition. This parameter adaptation prevents false positives caused by environmental variations while maintaining sensitivity to actual emergencies.
2Measurement precision
If sensor thresholds are set to be highly sensitive to detect all emergencies, then the detection precision is improved, but the number of false alarms increases
Solution Approach 1:
The system dynamically adjusts sensor thresholds rather than using fixed values. Thresholds are adapted in real-time based on the detected state of the environment and the person under care. During normal quiescent periods, thresholds operate at standard sensitivity levels, but when edge conditions are detected, thresholds are dynamically modified to prevent false alarms while maintaining detection precision for actual emergencies.
Solution Approach 2:
The system performs preliminary evaluation of sensor data against stored patterns in the elastic repository before triggering alerts. By pre-storing normal environmental patterns and sensor readings, the system can compare current readings against this baseline to determine if an alert is warranted. This preliminary action filters out false alarms before they are generated, maintaining high detection precision without excessive false positives.
3Reliability
If the system continuously monitors all sensor data to ensure accurate emergency detection, then the reliability is improved, but the energy consumption increases
Solution Approach 1:
The system implements periodic monitoring with variable intensity based on the detected state. During quiescent periods when no edge conditions are present, monitoring operates at reduced intensity with lower energy consumption. When edge conditions are detected, the system transitions to continuous high-intensity monitoring to ensure accurate emergency detection. This periodic action with adaptive intensity maintains monitoring reliability while significantly reducing overall energy consumption compared to continuous full-intensity monitoring.
Solution Approach 2:
The system uses the elastic repository to store and retrieve historical sensor patterns, enabling sensors to self-evaluate their data against stored patterns without requiring constant centralized processing. This self-service capability allows local evaluation and filtering of sensor data, reducing the energy required for data transmission and centralized analysis while maintaining reliable monitoring through distributed intelligence.
Data Source
AI summary
A system, apparatus, and method to monitor at least one person in at least one environment. The environment includes at least two sensors capable of detecting the presence of a person in that environment. The person under monitoring has a care condition to be monitored, where such monitoring involves the at least two sensors providing data sets to at least one signal monitoring system. Such data sets are communicated to at least one digital twin representing the person under monitoring and their environment, such that patterns of behavior may be determined for that person. Such patterns may be represented in the at least one digital twin, as to detect behavior that indicates a change in the care condition of that person under monitoring.