Cardiovascular Diseases Diagnosis Using an ECG Multi-Band Non-Linear Machine Learning Framework Analysis.

Area of Science:

• Biomedical Engineering
• Cardiology
• Machine Learning

Background:

• Cardiovascular diseases (CVDs) are the leading cause of global mortality.
• Accurate and early diagnosis of CVDs is crucial for effective treatment.
• Existing diagnostic methods can be invasive or lack comprehensive discriminatory power.

Purpose of the Study:

• To develop and validate a machine learning model for discriminating between seven specific CVDs and a healthy control group.
• To assess the efficacy of non-linear ECG features extracted using Discrete Wavelet Transform (DWT) for CVD classification.
• To compare the performance of the proposed method against existing studies using the PTB diagnostic ECG database.

Main Methods:

• Extracted 10 non-linear features from 1-second ECG lead signals using DWT multi-band analysis.
• Employed machine learning models trained and tested with a leave-one-out cross-validation approach.
• Performed binary and comprehensive all-vs-all classification analyses between CVDs and controls.

Main Results:

• Achieved accuracy rates ranging from 73% to 100% for discriminating between study groups.
• Reported Recall values between 68% and 100%.
• Area Under the Curve (AUC) results varied from 0.42 to 1, indicating strong discriminatory potential.

Conclusions:

• The proposed machine learning method effectively distinguishes between various cardiovascular diseases and healthy individuals.
• The approach offers advantages, including multi-class comparison and utilization of non-linear ECG analysis.
• This method represents a valuable tool for enhancing CVD diagnosis and patient stratification.