Physiological Change Detection Using Phase-Space Trajectory Matrices
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The medical community lacks accurate and portable equipment to detect early small events in physiological changes, such as neuronal firing changes prior to epileptic seizures, which often manifest long after the initial change.
Innovation Solution
A method involving embedding time series data into a reconstructed phase space, partitioning it into regions, and generating matrices based on transition probabilities and trajectory separation rates to detect anomalous conditions, with a processor configured to compare these matrices to identify state changes and potentially counteract them.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional detection methods are used, then equipment portability is maintained, but detection precision of early physiological changes deteriorates
Solution Approach 1:
The patent transforms 1D time series physiological data into 2D phase space representations through delay embedding, where each point is represented as (x(t), x(t+τ)). This dimensional transformation enables the detection of early physiological changes by visualizing trajectory patterns that are not apparent in the original time domain, thereby improving detection precision without requiring complex multi-sensor equipment
Solution Approach 2:
The phase space is partitioned into discrete regions or cells, and the system trajectory is tracked as it moves between these regions. By segmenting the continuous phase space into manageable regions and analyzing transition patterns between them, the method achieves high detection precision using simple computational operations that can be implemented in portable devices
2Measurement precision
If conventional detection methods are used, then time delay in detection is reduced, but detection precision of early small events deteriorates
Solution Approach 1:
The method establishes a baseline phase space trajectory pattern representing normal physiological conditions before actual events occur. By comparing real-time trajectories against this pre-established baseline, the system can detect early deviations that precede clinical events such as seizures or heart attacks, enabling early warning without time delay
Solution Approach 2:
The system continuously monitors phase space trajectories and compares them against expected patterns, providing real-time feedback on physiological state. When trajectory deviations exceed thresholds, the system generates alerts, creating a closed-loop detection mechanism that reduces time delay by immediately responding to early physiological changes rather than waiting for post-emergent analysis
3Loss of time
If post-emergent detection process is used, then detection simplicity is maintained, but detection timing deteriorates
Solution Approach 1:
The patent replaces complex post-emergent medical analysis systems with a simplified real-time computational approach using phase space reconstruction and trajectory analysis. By substituting mechanical/clinical detection processes with mathematical transformations of existing physiological signals, the system achieves early detection without requiring complex additional hardware or post-processing procedures
Data Source
AI summary
Systems and methods for detecting changes in a physiological system are provided. A method comprises embedding time series data relating to the physiological system within a reconstructed phase space and partitioning the reconstructed phase space into a plurality of regions. The method further comprises generating a first matrix having a plurality of cells. A value stored in each cell is based on a probability that the system will transition from a first region associated with the cell to a second region associated with the cell and a rate of separation of trajectories of the embedded data within at least one of the first region and the second region. The first matrix is associated with a normal operating condition of the physiological system. The method further comprises comparing the first matrix and a second matrix based on different data to detect an anomalous condition of the physiological system.