Dynamic Time Warping for Bio-Signal Waveform Accuracy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for determining representative waveforms of quasi-periodic bio-signals struggle with accuracy due to changing signal periods over time, affecting the reliability of bio-signal analysis in wearable medical devices.
Innovation Solution
A representative waveform providing apparatus and method utilizing a dynamic time warping (DTW) algorithm to segment periodic signals, determine similarity between segments, and select a representative waveform based on highest similarity or ensemble averaging, ensuring accurate representation of bio-signals with dynamic periods.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional waveform determination methods are used for quasi-periodic bio-signals, then the analysis process is simple, but the accuracy of bio-signal analysis deteriorates due to changing signal periods
Solution Approach 1:
The processor segments the periodic signal into multiple periodic signal segments based on detected periods. This segmentation allows the system to handle quasi-periodic signals with changing periods by analyzing each segment individually, thereby improving measurement precision without overwhelming complexity
Solution Approach 2:
The system dynamically adjusts to changing signal periods by detecting periods in real-time and segmenting signals accordingly. The dynamic time warping algorithm also dynamically aligns segments with varying periods, enabling accurate representation of quasi-periodic bio-signals while maintaining manageable complexity
2Reliability
If dynamic time warping algorithm is applied to determine representative waveform, then the accuracy of representing quasi-periodic signals is improved, but the computational complexity increases
Solution Approach 1:
By dividing the periodic signal into multiple segments before applying dynamic time warping, the computational complexity is reduced while maintaining reliability. Each segment is processed independently, making the overall system more manageable while still accurately representing quasi-periodic characteristics
Solution Approach 2:
The system applies dynamic time warping selectively to compare segments and determine representativeness, rather than processing the entire signal at once. This partial application of the algorithm maintains high reliability for quasi-periodic signal representation while controlling computational complexity
3Measurement precision
If signal segments with varying periods are compared directly, then the processing is straightforward, but the similarity determination accuracy deteriorates
Solution Approach 1:
The dynamic time warping algorithm dynamically adjusts the alignment of signal segments with varying periods, allowing accurate similarity determination. The system adapts to period changes in each segment, improving measurement precision while the modular processing approach keeps complexity manageable
Data Source
AI summary
A representative waveform providing apparatus and method, a blood pressure estimation apparatus, and a wearable device are provided. The representative waveform providing apparatus may include a signal obtainer configured to obtain a periodic signal and a processor configured to determine a representative waveform of the periodic signal by using a dynamic time warping algorithm.