Beat-Space ECG Analysis for Cardiovascular Death Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for predicting cardiovascular death following an acute cardiac event are limited by the ambiguity introduced by the quasi-periodicity of electrocardiogram (ECG) signals, particularly when analyzing frequency domains, as they can be either periodic with respect to time or heartbeats, leading to unclear observations across patients and over time.
Innovation Solution
A method that converts ECG time-series data from time-space to beat-space, computes power in various frequency bands, and uses a L1-regularized logistic regression machine learning program to generate a weighted risk vector for predicting patient outcomes, focusing on morphological and heart rate variability features.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If frequency domain analysis is performed on ECG signals in time-space, then periodic changes can be measured, but the quasi-periodicity of ECG signals introduces significant ambiguity and unclear observations across patients and over time
Solution Approach 1:
The patent transforms the reference frame from time-space to beat-space by changing the fundamental parameter from time (seconds) to heartbeat count. This parameter transformation resolves the quasi-periodicity problem by normalizing frequency measurements to beat intervals rather than time intervals, eliminating the ambiguity caused by variable heart rates across different patients and time points.
2Productivity
If traditional time-space frequency domain analysis is used, then computational methods are established, but the frequency bands measured differ between patients with different heart rates leading to ambiguous observations
Solution Approach 1:
Instead of measuring frequency in terms of time (traditional approach), the patent inverts the approach by measuring frequency in terms of heartbeat intervals. This inversion fundamentally changes the reference frame from temporal to event-based, ensuring that frequency bands are consistent across patients regardless of their heart rate, thereby resolving the ambiguity in frequency band measurements.
Data Source
AI summary
A method for the machine generation of a model for predicting patient outcome following the occurrence of an event. In one embodiment the method includes the steps of obtaining a physiological signal of interest, the physiological signal having a characteristic; obtaining a time series of a signal characteristic; dividing the time series into a plurality of window segments; converting the time series from time-space to beat-space; computing the power in various frequency bands of each window segment; computing the 90th percentile of the spectral energies across all window segments for each frequency band; and inputting the data into a machine learning program to generate a weighted risk vector.