Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Personalized Heart Disease Detection via ECG Digital Twin Generation

Yaojun Hu, Jintai Chen, Lianting Hu, Dantong Li, Jiahuan Yan, Haochao Ying, Huiying Liang, Jian Wu

TL;DR

An innovative prospective learning approach is presented, which generates digital twins of healthy individuals' anomalous ECGs and enhances the model sensitivity to the personalized symptoms and ensures robust privacy protection, safeguarding patient data in model development.

Abstract

Heart diseases rank among the leading causes of global mortality, demonstrating a crucial need for early diagnosis and intervention. Most traditional electrocardiogram (ECG) based automated diagnosis methods are trained at population level, neglecting the customization of personalized ECGs to enhance individual healthcare management. A potential solution to address this limitation is to employ digital twins to simulate symptoms of diseases in real patients. In this paper, we present an innovative prospective learning approach for personalized heart disease detection, which generates digital twins of healthy individuals' anomalous ECGs and enhances the model sensitivity to the personalized symptoms. In our approach, a vector quantized feature separator is proposed to locate and isolate the disease symptom and normal segments in ECG signals with ECG report guidance. Thus, the ECG digital twins can simulate specific heart diseases used to train a personalized heart disease detection model. Experiments demonstrate that our approach not only excels in generating high-fidelity ECG signals but also improves personalized heart disease detection. Moreover, our approach ensures robust privacy protection, safeguarding patient data in model development.

Personalized Heart Disease Detection via ECG Digital Twin Generation

TL;DR

An innovative prospective learning approach is presented, which generates digital twins of healthy individuals' anomalous ECGs and enhances the model sensitivity to the personalized symptoms and ensures robust privacy protection, safeguarding patient data in model development.

Abstract

Heart diseases rank among the leading causes of global mortality, demonstrating a crucial need for early diagnosis and intervention. Most traditional electrocardiogram (ECG) based automated diagnosis methods are trained at population level, neglecting the customization of personalized ECGs to enhance individual healthcare management. A potential solution to address this limitation is to employ digital twins to simulate symptoms of diseases in real patients. In this paper, we present an innovative prospective learning approach for personalized heart disease detection, which generates digital twins of healthy individuals' anomalous ECGs and enhances the model sensitivity to the personalized symptoms. In our approach, a vector quantized feature separator is proposed to locate and isolate the disease symptom and normal segments in ECG signals with ECG report guidance. Thus, the ECG digital twins can simulate specific heart diseases used to train a personalized heart disease detection model. Experiments demonstrate that our approach not only excels in generating high-fidelity ECG signals but also improves personalized heart disease detection. Moreover, our approach ensures robust privacy protection, safeguarding patient data in model development.
Paper Structure (4 sections, 5 figures, 1 algorithm)

This paper contains 4 sections, 5 figures, 1 algorithm.

Figures (5)

  • Figure 1: Model performance with different values of the threshold, $l$. DTW: Dynamic Time Warping; ATE: Average Treatment Effect; ECG Wise ACC: ECG Wise Accuracy.
  • Figure 2: Example of ECG described as "sinus tachycardia apart from rate, normal ecg." generated from the real normal ECG.
  • Figure 3: Visualization of generated ECGs with the Hypertrophy (HYP) heart disease patterns by different models. For better viewing, we only display three leads of ECGs: I, aVR, and V3.
  • Figure 4: Visualization of generated ECGs with the Myocardial Infarction (MI) heart disease patterns by different models. For better viewing, we only display three leads of ECGs: I, aVR, and V3.
  • Figure 5: Visualization of generated ECGs with the Conduction Disturbance (CD) heart disease patterns by different models. For better viewing, we only display three leads of ECGs: I, aVR, and V3.