Closed-Loop Wearable Scratch Detection With On-Device ML Feedback
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing technologies fail to accurately, objectively, and continuously assess itch by quantifying scratch or other physiological symptoms in a user's natural environment, leading to limited therapeutic options and delays in analysis due to remote data communication.
Innovation Solution
A wearable sensor device with an inertial measurement unit and microcontroller unit that detects motion and acoustic signatures, performs machine learning operations to predict scratching activity, and provides haptic feedback to the user.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Power
If sensor data is communicated to a remote location for analysis, then analysis can be performed with sufficient computational resources, but this contributes to lag and delay in analysis performance
Solution Approach 1:
The patent introduces an on-device machine learning service as an intermediary between the sensor data and remote analysis. This service processes data locally on the wearable device, filtering and pre-processing information before transmitting only relevant results to remote systems, thereby reducing transmission delays and bandwidth requirements while maintaining analytical capability
Solution Approach 2:
The patent segments the analytical processing into two parts: time-critical processing performed locally on the wearable device using machine learning models, and non-time-critical processing performed remotely. This segmentation allows immediate detection and response to scratching events while still utilizing remote computational resources for comprehensive analysis
2Measurement precision
If existing techniques are used to assess itch, then some data can be collected, but they fail to accurately, objectively, and continuously assess itch by quantifying scratch in a user's natural environment
Solution Approach 1:
The patent implements a multi-functional sensor system that simultaneously performs motion detection, acoustic signature analysis, and physiological parameter monitoring. This universal approach enables accurate scratching detection across diverse natural environments without requiring environment-specific calibration, as the system adapts to various contexts through machine learning
Solution Approach 2:
The patent employs closed-loop feedback where the machine learning service continuously analyzes sensor data, provides real-time detection results, and adjusts processing parameters based on detected patterns. This feedback mechanism enables continuous, objective assessment of scratching behavior in natural environments with high temporal resolution
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 real-time, accurate detection and mitigation of scratching activity, improving sleep quality and reducing the itch-scratch cycle through in-sensor computation and haptic feedback.
Implementation Method 1
an inertial measurement unit (IMU) for detecting the motion and acoustic signatures
Implementation Method 2
an inertial measurement unit (IMU) for detecting the motion and acoustic signatures
Implementation Method 3
the machine-learning service predicts the detection of human scratching activity by performing a first machine-learning operation on the feature-related data
Implementation Method 4
a vibratory motor for alerting the human subject
Data Source
AI summary
Methods and electronic devices for measuring motion and acoustic signatures of physiological processes of a human subject. The method includes measuring motion and acoustic signatures of physiological processes of a human subject; sending the first feature-related data to a machine learning service; sending the first feature-related data to a machine learning service; and determining a predicted detection of human scratching activity by the machine-learning service by performing a machine-learning operation on the feature-related data.