ASD Detection via Threshold-Based Sensor Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing sensor-based systems for detecting Autism Spectrum Disorder (ASD) are inefficient and inaccurate, often requiring expensive equipment and causing battery drain in smartphones, making them impractical for home use and inefficient in processing operations.
Innovation Solution
A portable device system that processes heart rate, accelerometer, and audio data to detect ASD episodes by activating data collection and feature extraction based on predetermined thresholds, transmitting alerts when specific conditions are met, thereby improving detection efficiency and accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If existing sensor-based systems are used to detect ASD, then detection capability is provided, but device complexity and cost increase due to expensive sensors and equipment
Solution Approach 1:
The patent segments the detection process into multiple stages with different sensor requirements. Stage 1 uses simple heart rate monitoring to filter obvious non-ASD cases. Stage 2 activates only when needed, using accelerometer and audio sensors for detailed analysis. This segmentation allows the system to achieve reliable ASD detection while avoiding the need for all sensors to operate simultaneously, thereby reducing overall device complexity.
Solution Approach 2:
The patent implements dynamic sensor activation based on real-time conditions. Sensors are selectively activated or deactivated depending on the current detection stage and heart rate thresholds. This dynamic approach allows the system to maintain high detection reliability when needed while minimizing device complexity during normal operation, as not all sensors must be present or active at once.
2Reliability
If smart phones are used to process sensor data for ASD detection, then detection function is provided, but computational efficiency decreases and battery drain increases
Solution Approach 1:
The patent segments sensor data processing into distinct stages with different computational requirements. Stage 1 performs simple heart rate threshold comparisons that consume minimal energy. Stage 2 performs more computationally intensive accelerometer and audio analysis only when Stage 1 triggers, thereby reducing overall energy consumption while maintaining detection reliability.
Solution Approach 2:
The patent implements periodic sensor activation based on heart rate monitoring. Sensors are activated periodically only when heart rate exceeds predetermined thresholds, rather than continuously. This periodic action significantly reduces battery drain while maintaining the ability to detect ASD episodes when they occur.
3Measurement precision
If continuous sensor data collection is performed, then detection accuracy is improved, but battery strain increases
Solution Approach 1:
The patent segments data collection into two phases: continuous heart rate monitoring (low energy) and conditional accelerometer/audio collection (high energy but infrequent). This segmentation allows the system to maintain detection accuracy by collecting comprehensive data when needed while minimizing battery strain by avoiding continuous high-energy sensor operation.
Solution Approach 2:
The patent performs preliminary heart rate monitoring and threshold comparison before activating more energy-intensive sensors. This preliminary action filters out cases that don't meet ASD criteria, allowing the system to maintain high detection accuracy for actual ASD episodes while reducing overall battery strain by avoiding unnecessary sensor activation.
Data Source
AI summary
Technologies and techniques for detecting and alerting onset Autism Spectrum Disorder (ASD), where a device receives sensor data including heart rate data, accelerometer data and audio data. The device is configured to activate the collection of accelerometer data and extract accelerometer data features comprising frequency components if the heart rate exceeds a first threshold, activate the collection of audio data and extract audio data features comprising frequency components if the heart rate data meets or exceeds a second predetermined threshold, determine if the audio data meets or exceeds a predetermined audio threshold comprising frequency characteristics, and transmit an alert indicating the detection of an onset ASD episode, based on the extracted audio data features, the extracted accelerometer data features and the heart rate data.