Heart Rate Variability Pattern Analysis via Multidimensional Phase Space
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods fail to provide a compact, universal, and adaptable representation of cyclic or quasi-periodical dissipative systems, particularly for heart rate variability analysis, which is essential for diagnosing and prognosing cardiac health, as they either overlap time scales or conceal complex multidimensional sequences, making it difficult to identify universal variability patterns and pathological arrhythmias.
Innovation Solution
A system that calculates and visually represents the normalized variability of time intervals between heart cycles using a specific algorithmic transformation, allowing for spatial representation in multiple dimensions, including color, to identify common patterns and correlations, and compare individual systems with universal patterns, independent of time scales.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If Fourier transform or Wavelet transforms are used for time scale analysis, then frequency domain information is obtained, but time scales overlap and universal variability patterns become inaccessible
Solution Approach 1:
The patent transforms the one-dimensional time series of heart rate intervals into N-dimensional space (where N≥2), creating a multidimensional phase space representation. This dimensional transformation allows separation of overlapping time scales that cannot be distinguished in traditional frequency domain analysis, making universal variability patterns accessible without the limitations of Fourier or Wavelet transforms.
Solution Approach 2:
The patent segments the continuous time series into discrete vectors of N consecutive time intervals, creating N-dimensional vectors that represent local temporal patterns. This segmentation approach allows analysis of variability at multiple scales simultaneously without the overlapping problems of traditional transforms, as each vector captures a specific temporal window's characteristics.
2Device complexity
If traditional HRV analysis methods are used, then time domain parameters are calculated, but complex multidimensional sequences are concealed and compact representation is not achieved
Solution Approach 1:
The patent merges multiple time domain parameters (mean, standard deviation, skewness, kurtosis, etc.) into a single N-dimensional vector for each window of consecutive intervals. This consolidation provides a compact representation that preserves all multidimensional sequence information, allowing comprehensive characterization of heart rate variability in a unified structure that is both compact and information-rich.
3Measurement precision
If high dimensional representation is used to capture complex patterns, then identification accuracy improves, but visualization and interpretation become difficult
Solution Approach 1:
The patent applies different analytical treatments to different aspects of the N-dimensional vectors. Statistical parameters (mean, standard deviation, skewness, kurtosis) are calculated to characterize local temporal patterns within each window, while the full N-dimensional vectors preserve global sequence information. This local quality approach enables both high-dimensional pattern recognition and simplified statistical interpretation.
Solution Approach 2:
The patent uses color coding to represent different dimensions or characteristics of the N-dimensional vectors in visualizations. By mapping high-dimensional data to visual attributes like color, the system enables intuitive interpretation of complex patterns while maintaining the full dimensional information for accurate pattern identification and comparison with reference patterns.
Data Source
Figure 1
Figure 2(a)~2(d)
Figure 3(a)~3(b3)
AI summary
The invention relates to a method for providing a description, graphical representation, and a graphical identification of specific operating patterns of quasi-periodic cyclic systems, such as, but not limited to, reciprocating combustion engines, rotary machines, or biological organs such as the heart. The invention also relates to a method for calculating an indicator evaluating the heart health or condition of an individual, as well as for diagnosing and issuing prognoses relating to the functionality, pathology, or standard of health of a machine or organism equipped with a motor or organ that operates cyclically, and to provide a description, a compact graphical representation, and a graphical identification of specific operating patterns of dynamic systems, e.g. economic systems such as the stock market.