Non-Contact Sleep Stage Classification Using RF and Pressure Sensors
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for determining human sleep stages are cumbersome, require direct electrical or mechanical contact, and have limited accuracy in distinguishing between sleep stages, especially for patients with sleep disorders.
Innovation Solution
A non-contact system using radio-frequency motion sensors and pressure-sensitive mattresses to measure bodily movement and respiration, analyzing variability and amplitude to classify sleep stages without the need for ECG or bioimpedance signals, providing feedback on sleep quality and stage.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If polysomnography with multiple EEG readings and contact sensors is used, then measurement precision of sleep stages is improved, but device complexity and ease of operation deteriorate
Solution Approach 1:
The patent extracts and eliminates the need for complex polysomnography equipment (multiple EEG channels, EOG, ECG sensors) while retaining the core functionality of sleep stage classification. It uses only a single EEG channel in combination with movement and respiration signals to achieve accurate sleep stage determination, thereby reducing device complexity while maintaining measurement precision.
Solution Approach 2:
The system achieves multi-functionality by using a minimal sensor set (single EEG channel, movement sensor, respiration sensor) to perform multiple sleep stage classification functions. The same simplified sensor configuration can distinguish between wake, N1, N2, N3, and REM stages, as well as detect sleep disorders, eliminating the need for separate specialized equipment for each measurement type.
2Measurement precision
If polysomnography with multiple contact sensors is used, then measurement precision is improved, but ease of operation deteriorates
Solution Approach 1:
The patent removes the requirement for multiple contact sensors and complex electrode placements while maintaining sleep stage classification accuracy. It achieves this by extracting only the essential signals needed (single EEG channel, movement, and respiration) and using advanced signal processing to compensate for the reduced sensor input, making the system much easier to operate.
Solution Approach 2:
The system incorporates automated signal processing and classification algorithms that automatically analyze the minimal sensor data without requiring manual intervention or expert interpretation. The processor automatically distinguishes sleep stages and detects abnormalities, making the system easy to operate while maintaining high measurement precision.
3Ease of operation
If actigraphy with accelerometers is used, then ease of operation is improved, but measurement precision deteriorates
Solution Approach 1:
The patent combines movement sensing (similar to actigraphy) with respiration rate analysis and single-channel EEG processing to achieve accurate sleep stage differentiation. By merging these multiple signal types and analyzing them together through sophisticated signal processing, the system achieves measurement precision comparable to full polysomnography while maintaining the ease of operation of wearable devices.
Solution Approach 2:
The system analyzes multiple parameters from each signal type (movement amplitude, respiration rate, EEG frequency content) and combines them to determine sleep stage. By changing from simple movement counting to multi-parameter analysis, the system achieves high measurement precision while keeping the device simple and easy to operate.
4Measurement precision
If full polysomnography signals are measured, then sleep stage differentiation accuracy is improved, but loss of time for implementation increases
Solution Approach 1:
The patent extracts and eliminates the time-consuming aspects of full polysomnography setup (multiple electrode placements, complex equipment configuration) while retaining the essential measurement capability. The simplified sensor setup can be applied quickly, and the automated processing provides immediate results, reducing implementation time while maintaining sleep stage differentiation accuracy.
Solution Approach 2:
The system performs preliminary signal processing and feature extraction automatically as data is collected, preparing the data for classification in real-time. This preliminary action eliminates the need for time-consuming manual analysis and allows for rapid implementation and immediate interpretation of sleep stages.
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 accurate differentiation between deep sleep, light sleep, and REM sleep without direct contact, offering real-time and post-sleep feedback for improved sleep assessment and quality metrics.
Implementation Method 1
one or more sensors configured to receive a reflected radio-frequency signal off a living subject
Data Source
AI summary
Methods and apparatus monitor health by detection of sleep stage. For example, a sleep stage monitor may access sensor data signals related to bodily movement and respiration movements. At least a portion of the detected signals may be analyzed to calculate respiration variability. The respiration variability may include variability of respiration rate or variability of respiration amplitude. A processor may then determine a sleep stage based on a combination bodily movement and respiration variability. The determination of sleep stages may distinguish between deep sleep and other stages of sleep, or may differentiate between deep sleep, light sleep and REM sleep. The bodily movement and respiration movement signals may be derived from one or more sensors, such as non-invasive sensor (e.g., a non-contact radio-frequency motion sensor or a pressure sensitive mattress).