Sleep Stage Classification Using Circadian Rhythm Adjustment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional sleep detection and classification techniques implemented by wearable devices do not consider circadian rhythm adjustment, leading to deficient sleep stage classification.
Innovation Solution
A system that utilizes a machine learning classifier with circadian rhythm-derived features to classify sleep stages, tailoring algorithms to individual users by adjusting physiological data based on their unique circadian rhythms.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional sleep detection techniques are used, then device complexity is reduced, but sleep stage classification accuracy deteriorates
Solution Approach 1:
The patent transforms raw physiological signals into circadian rhythm-derived features by adjusting parameters based on time-of-day and individual circadian patterns. This parameter transformation enables the machine learning classifier to capture temporal variations in sleep stages, improving classification accuracy without requiring fundamentally more complex hardware
Solution Approach 2:
The patent introduces circadian rhythm-derived features as an intermediary between raw physiological data and sleep stage classification. These features act as a bridge that encapsulates temporal information and individual circadian patterns, allowing the machine learning model to achieve higher accuracy while maintaining manageable complexity through feature engineering rather than model complexity
2Measurement precision
If circadian rhythm adjustment models are incorporated, then sleep stage classification accuracy is improved, but processing time increases
Solution Approach 1:
The patent performs preliminary computation of circadian rhythm-derived features during data collection and preprocessing stages. By pre-computing these features based on time-of-day and individual circadian patterns before final classification, the system reduces the computational burden during real-time sleep stage determination, thereby decreasing overall processing time while maintaining accuracy
Solution Approach 2:
The patent segments the processing pipeline into distinct stages: data collection, circadian feature computation, and sleep stage classification. This segmentation allows each component to be optimized independently, with circadian features pre-computed and stored for efficient retrieval during classification, reducing the time required for the most computationally intensive operations
3Measurement precision
If personalized circadian rhythm tailoring is implemented, then sleep analysis accuracy is improved, but device complexity increases
Solution Approach 1:
The patent implements a feedback mechanism where individual user data is continuously used to refine and update personal circadian rhythm models. By incorporating feedback from repeated measurements and user-specific patterns, the system achieves increasing accuracy over time while managing complexity through adaptive learning rather than requiring overly complex fixed algorithms
Solution Approach 2:
The patent employs dynamic adjustment of circadian rhythm parameters based on individual user responses and patterns. Rather than using static, one-size-fits-all algorithms, the system dynamically adapts to each user's unique circadian characteristics, achieving personalized accuracy while controlling complexity through flexible parameter adjustment rather than complex fixed rules
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Methods, systems, and devices for sleep staging algorithms are described. A system may receive physiological data associated with a user from a wearable device, where the physiological data may be collected via the wearable device throughout a time interval. The system may identify a circadian rhythm adjustment model configured to weight the physiological data based on a circadian rhythm associated with the user. The system may input the physiological data and the circadian rhythm adjustment model into a machine learning classifier, and classify the physiological data, using the machine learning classifier, into at least one sleep stage of a set of sleep stages for at least a portion of the time interval, where the classifying is based on the circadian rhythm adjustment model. A graphical user interface (GUI) of a user device may display an indication of the at least one sleep stage based on classifying the physiological data.