Sleep Stage Estimation Using Pulse and Body Movement Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing sleep state monitoring systems face challenges in accurately determining sleep stages using brain waves, body movements, and pulse waves, particularly in terms of efficiency and ease of data acquisition.
Innovation Solution
A sleep stage estimation device that utilizes pulsation data and body movement data to estimate sleep stages through a combination of logistic regression models for sleep stage probabilities, sigmoid functions for stage transitions, and conditional random fields to integrate stage probabilities and transition probabilities.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If brain waves are used to determine sleep stages, then measurement precision is improved, but device complexity and difficulty of data acquisition increase
Solution Approach 1:
The patent extracts and removes the need for complex brain wave acquisition systems by focusing only on pulse wave and body movement data. This extraction principle eliminates the harmful complexity while preserving the essential sleep stage determination function through alternative, simpler physiological parameters.
Solution Approach 2:
The patent uses pulse waves and body movements as surrogate indicators that copy the information contained in brain waves regarding sleep stages. Instead of directly measuring brain activity, the system captures equivalent sleep state information through these accessible physiological parameters, achieving the same diagnostic purpose with simpler means.
2Measurement precision
If brain waves are used to determine sleep stages, then measurement precision is improved, but ease of operation deteriorates
Solution Approach 1:
The patent extracts the essential sleep stage information from complex brain wave measurements and represents it through simple pulse wave and body movement data. This extraction makes the system easier to operate by eliminating the need for complex EEG equipment and procedures while maintaining diagnostic accuracy.
Solution Approach 2:
The patent employs pulse waves and body movements as accessible proxies that copy sleep stage information. These parameters are much easier to measure and acquire in practice compared to brain waves, improving operational ease while preserving the ability to accurately determine sleep stages.
3Ease of operation
If pulse waves and body movements are used instead of brain waves, then ease of operation is improved, but measurement precision deteriorates
Solution Approach 1:
The patent merges multiple simple parameters (pulse wave characteristics and body movement data) into a unified analysis framework. By combining these easily acquired parameters through conditional random fields and logistic regression, the system achieves measurement precision comparable to brain wave analysis while maintaining operational simplicity.
Solution Approach 2:
The patent creates a composite analytical model that integrates pulse wave data and body movement data. This composite approach combines the strengths of multiple simple parameters to achieve accurate sleep stage determination, equivalent to using complex brain wave data alone, while preserving the ease of data collection.
Data Source
Figure 1
Figure 2~3
Figure 4
AI summary
One embodiment of the present invention pertains to a sleep stage estimation device that is provided with: a test subject data acquisition unit which acquires pulsation data and body movement data of a test subject; a sleep stage probability estimation unit which acquires a feature quantity sequence from the pulsation data and estimates, using a learned sleep stage probability estimation model, a sleep stage probability sequence of the test subject from the acquired feature quantity sequence; a sleep stage transition probability estimation unit which acquires a body movement amount sequence from the body movement data and estimates, by using a learned sleep stage transition probability estimation model, a sleep stage transition probability sequence of the test subject from the acquired body movement amount sequence; and a sleep stage estimation unit which estimates, by using a learned conditional random field model, a sleep stage sequence of the test subject from the sleep stage probability sequence and the sleep stage transition probability sequence.