Contactless Event Monitoring With AI-Based Early Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
There is a need for continuous, contactless monitoring of individuals, particularly those who are sick, elderly, or have medical conditions, to detect undesired events such as falls, seizures, or abnormal breathing, without privacy intrusion and with minimal caretaker involvement, especially in scenarios like pandemics or isolated environments.
Innovation Solution
A system utilizing data aggregation devices, signal processors, and artificial intelligence algorithms to predict undesired events by analyzing collected data, implementing deep learning and rule-based engines to determine thresholds for alerts, and correcting false predictions through self-healing mechanisms.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If continuous monitoring is performed using traditional contactless methods, then monitoring coverage is improved, but false alerts increase and resource consumption increases
Solution Approach 1:
The system dynamically adjusts monitoring parameters such as detection thresholds, sampling rates, and alert sensitivity based on baseline measurements of each individual's normal behavior patterns. This allows the system to maintain high monitoring coverage while reducing false alerts and optimizing resource usage by adapting to each person's unique characteristics.
Solution Approach 2:
The system performs preliminary baseline measurements during an initial period to establish normal behavior patterns before full monitoring begins. This preliminary action enables the system to distinguish between normal variations and actual anomalies, reducing false alerts while maintaining comprehensive monitoring coverage from the start.
2Reliability
If traditional contactless monitoring methods are used, then monitoring capability is improved, but privacy intrusion increases
Solution Approach 1:
The system extracts only the essential monitoring data needed for detection purposes while deliberately excluding personal identifiable information and sensitive private data. By taking out only the necessary parameters (movement patterns, physiological signals) and leaving out extraneous personal information, the system maintains monitoring capability while minimizing privacy intrusion.
Solution Approach 2:
The system uses intermediary processing layers that aggregate and anonymize data before storage and analysis. Personal data is transformed into standardized, non-identifiable metrics that preserve monitoring accuracy while preventing direct identification of individuals, thus acting as a mediator between monitoring needs and privacy protection.
3Measurement precision
If frequent caretaker visits are implemented for monitoring, then detection accuracy is improved, but cost and impracticality increase
Solution Approach 1:
The system replaces the mechanical system of manual caretaker monitoring with automated sensor-based detection and AI analysis. Sensors continuously collect physiological and behavioral data, while algorithms automatically analyze patterns and detect anomalies, achieving high detection accuracy without requiring physical caretaker presence or manual intervention.
Solution Approach 2:
The monitoring system performs self-assessment by automatically collecting, analyzing, and interpreting data without external human intervention. The AI algorithms continuously learn from incoming data and adjust detection parameters autonomously, enabling the system to maintain high detection accuracy while eliminating the need for dedicated caretaker resources.
Data Source
AI summary
Disclosed is a contactless event monitoring and early detection method and system of real time monitoring of a person to predict an undesired event. The contactless event monitoring and early detection method collects data from a one or more data aggregation devices and filters the collected data in a signal processor. The contactless event monitoring and early detection method then analyzes the collected data in an event monitoring and detection processor. The event monitoring and detection processor implementing artificial intelligence algorithms to predict the probability of the occurrence of to the undesired event. Further, the event monitoring and detection processor receives a health data related to the person and selects an analytical model to be applied for prediction of the undesired event. In one embodiment, the probability of occurrence of the undesired event is associated on a threshold value. The threshold value is related to at least one early indicator associated with the person. The contactless event monitoring and early detection method then preempts the person about the onset of the undesired event by sending an alert to the person and a caretaker for remedial action.