Wearable Sensor Data Extraction for Fall Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current personal emergency response systems rely on complex central monitoring units that lack the capacity for effective communication of user data for post-mortem analysis and ongoing improvements, particularly in detecting falls and monitoring physiological characteristics without manual activation.
Innovation Solution
A personal monitoring system that includes a wearable device capable of sensing and recording physiological, motion, and location data, transmitting this data over a network to a computing device for real-time or subsequent access, and automatically notifying caregivers or emergency services in case of an emergency, such as a fall, with the ability to improve algorithms over time without capturing entire sampled data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If complex central monitoring units are used for fall detection and physiological monitoring, then detection capability is improved, but device complexity increases and data communication capacity is insufficient for post-mortem analysis
Solution Approach 1:
The patent extracts the data processing and analysis functions from the complex central monitoring unit and relocates them to the wearable device itself. The wearable device now performs event detection, feature extraction, and stores relevant data locally, eliminating the need for complex centralized processing while maintaining high detection capability and enabling post-mortem analysis of individual events.
2Measurement precision
If entire sampled data is captured for analysis, then measurement accuracy is improved, but data storage requirements and transmission bandwidth increase significantly
Solution Approach 1:
The patent extracts only the most relevant features and event information from the raw sampled data using feature extraction algorithms. Instead of transmitting and storing entire datasets, the system extracts and transmits only the essential event parameters, characteristics, and processed information, dramatically reducing data volume while preserving measurement accuracy for post-mortem analysis.
Solution Approach 2:
The system uses feedback from real-time event detection and feature extraction to dynamically adjust data collection and transmission strategies. By continuously analyzing detected events and their characteristics, the system optimizes which data to capture and transmit, ensuring high analytical accuracy while minimizing data quantity.
3Loss of time
If real-time emergency response is implemented, then response time is improved, but computational processing requirements increase
Solution Approach 1:
The patent segments the computational processing into two distinct locations: the wearable device handles real-time event detection, feature extraction, and immediate response decisions to enable fast emergency response; the central server performs comprehensive post-mortem analysis of extracted features. This segmentation allows real-time processing with limited computational power at the wearable while maintaining comprehensive analysis capability.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Enables timely and effective emergency response by providing real-time user data for immediate action, facilitating post-event analysis for system improvements, and enhancing fall detection capabilities without manual activation.
Implementation Method 1
an accelerometer for obtaining motion data indicative of movement of the user
Implementation Method 2
a barometer for obtaining data indicative of an elevation of the user
Data Source
AI summary
A personal monitoring system of the present disclosure has a network and a sensing device. The sensing device has a network interface for coupling the sensing device to the network and is coupled to a user for sensing raw data at a discrete time related to the user. Additionally, the system has logic that associates a timestamp with the raw data at the discrete time and stores the raw data as raw history data indicative of a plurality of raw data from discrete times. Further, the logic determines, based upon the raw history data, whether an event has occurred.