Phase Space ECG Analysis for Non-Invasive Pulmonary Hypertension Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for diagnosing pulmonary hypertension, particularly pulmonary arterial hypertension (PAH), are invasive and lack accuracy, relying on invasive procedures like right heart catheterization for confirmation, which is not feasible for reliable non-invasive diagnosis.
Innovation Solution
Utilizes phase space tomography and machine learning to generate tomographic images and mathematical features from biophysical signals, employing a trained neural network classifier to assess the presence of pulmonary hypertension non-invasively, analyzing electrical conduction patterns and geometric characteristics of the heart.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If invasive right heart catheterization is used to diagnose pulmonary hypertension, then measurement precision is improved, but ease of operation deteriorates and loss of time increases
Solution Approach 1:
The patent replaces the mechanical invasive catheterization system with a non-invasive electrical signal analysis system. ECG signals are processed through phase space transformation and machine learning algorithms to diagnose pulmonary hypertension, eliminating the need for physical catheter insertion while maintaining diagnostic capability
Solution Approach 2:
The patent introduces an intermediary computational system that processes ECG signals through phase space transformation and neural network classification. This intermediary layer translates routine ECG data into diagnostic information for pulmonary hypertension, bridging the gap between non-invasive measurement and accurate diagnosis
2Reliability
If invasive right heart catheterization is performed for confirmation, then reliability is improved, but productivity deteriorates
Solution Approach 1:
The patent performs preliminary diagnostic assessment using routine ECG signals before catheterization would be considered. By analyzing phase space features and applying machine learning classification to ECG data, the system provides early diagnostic indication that can confirm or rule out pulmonary hypertension without proceeding to invasive procedures
Solution Approach 2:
The patent creates a computational model that replicates the diagnostic function of catheterization using readily available ECG signals. The phase space transformation and neural network classifier produce a virtual diagnostic assessment that mirrors the information obtained from invasive measurement, eliminating the need for physical catheter insertion in many cases
3Ease of operation
If phase space tomography and machine learning are used for non-invasive diagnosis, then ease of operation is improved, but measurement precision may deteriorate
Solution Approach 1:
The patent transforms one-dimensional ECG time series signals into multi-dimensional phase space representations. By constructing phase portraits with multiple dimensions (e.g., x, y, z coordinates in 3D phase space), the system extracts additional diagnostic information from routine signals, enhancing measurement precision while maintaining non-invasive operation
Solution Approach 2:
The patent changes the parameter representation of ECG signals by applying phase space transformation and extracting geometric features (area, perimeter, fractal dimension). These transformed parameters provide new diagnostic metrics that improve measurement precision while keeping the original ECG recording non-invasive
Data Source
AI summary
Phase space tomography methods and systems to facilitate the analysis and evaluation of complex, quasi-periodic system by generating computed phase-space tomographic images and mathematical features as a representation of the dynamics of the quasi-periodic cardiac systems. The computed phase-space tomographic images can be presented to a physician to assist in the assessment of presence or non-presence of disease. In some implementations, the phase space tomographic images are used as input to a trained neural network classifier configured to assess for presence or non-presence of pulmonary hypertension, including pulmonary arterial hypertension.