Physiological Data Windowing for Continuous Human State Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for detecting and monitoring human states, such as drowsiness and cognitive load, face challenges due to the dynamic and variable nature of physiological manifestations, leading to inaccurate predictions and resource-intensive data collection, with existing datasets failing to capture the full range of manifestations and requiring inefficient resource utilization.
Innovation Solution
A method and system for recording physiological data over a continuous duration, segmenting it into shorter analysis windows, and labeling each with a ground truth label to create a more representative dataset for training accurate human state prediction models, leveraging extended-duration recordings to capture a broader spectrum of physiological responses.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If longer recording windows are used to capture full physiological dynamics, then measurement precision is improved, but time resolution for continuous monitoring deteriorates
Solution Approach 1:
The patent segments long continuous physiological recordings into multiple shorter analysis windows, each processed independently to generate state predictions. This allows the model to use sufficient data length for accurate state detection while providing frequent predictions across the entire recording period, effectively resolving the contradiction between measurement precision and time resolution.
2Loss of time
If shorter analysis windows are used for continuous monitoring, then time resolution is improved, but measurement precision deteriorates
Solution Approach 1:
The patent performs preliminary action by collecting and storing long continuous physiological recordings before the actual state detection process. These pre-collected long-duration datasets serve as the foundation for generating multiple shorter analysis windows, ensuring that each window contains sufficient physiological dynamics for accurate state detection while enabling frequent predictions.
3Measurement precision
If extended-duration recordings are used to capture full range of manifestations, then measurement precision is improved, but resource utilization deteriorates
Solution Approach 1:
The patent segments extended-duration recordings into multiple shorter analysis windows that can be processed independently and in parallel. This segmentation enables efficient data processing by distributing the computational load across multiple windows, improving productivity while maintaining the completeness of state representation through the use of sufficiently long original recordings.
Solution Approach 2:
The patent applies partial action by using a threshold duration that is sufficient to capture the full range of physiological manifestations needed for accurate state detection. Rather than indefinitely extending recording durations, the method identifies and uses the minimum necessary duration (threshold) to achieve complete state representation, optimizing resource utilization.
4Measurement precision
If threshold duration is set to capture complete physiological dynamics, then measurement precision is improved, but loss of time increases
Solution Approach 1:
The patent segments the continuous physiological recording at the threshold duration into multiple overlapping or adjacent analysis windows. Each window is processed to generate a state prediction, and these predictions are temporally aligned to provide continuous monitoring. This approach ensures that each prediction is based on sufficient data (threshold duration) while eliminating prediction delays through continuous segment processing.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The present disclosure relates to methods and systems for establishing ground truth in human state detection. In one embodiment, the disclosure teaches recording physiological data over a continuous duration of at least a threshold duration that is sufficient for capturing representation of the target state in bio signals used for detection. The recorded data is segmented into shorter analysis windows, as a step toward continuous state detection, each window less than the threshold duration, and labeled with a ground truth indicative of the target state. The windows and labels are stored in non-transitory memory, for later use in training, testing, and validating one or more state prediction models. The method enhances the accuracy of state detection models by providing a more representative ground truth through prolonged recordings, which capture the variable manifestations of human states in physiological processes.