ENACT-Heart -- ENsemble-based Assessment Using CNN and Transformer on Heart Sounds

2025-02-24

AI Summary

To address the limitation of insufficient heart sound classification accuracy in cardiovascular intelligent diagnosis, this paper proposes a Mixture-of-Experts (MoE) ensemble framework integrating Convolutional Neural Networks (CNNs) and Vision Transformers (ViTs). It is the first work to establish a CNN–ViT collaborative MoE paradigm for heart sound analysis, effectively leveraging the complementary strengths of CNNs in capturing local time-frequency patterns and ViTs in modeling long-range temporal dependencies. Evaluated on the standard four-class heart sound classification task, the proposed model achieves 97.52% accuracy—outperforming standalone CNN (95.45%) and ViT (93.88%) baselines by 2.07% and 3.64%, respectively. This work introduces a novel multimodal modeling approach for bioacoustic signals and advances the development of interpretable, robust automated cardiovascular health monitoring systems.

Technology Category

• Machine Learning: Mixture of Experts (MoE)
• Computer Vision: Multi-modal Vision
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Search and Retrieval-Augmented AI: Multilingual and cross-lingual Web search

Abstract

This study explores the application of Vision Transformer (ViT) principles in audio analysis, specifically focusing on heart sounds. This paper introduces ENACT-Heart - a novel ensemble approach that leverages the complementary strengths of Convolutional Neural Networks (CNN) and ViT through a Mixture of Experts (MoE) framework, achieving a remarkable classification accuracy of 97.52%. This outperforms the individual contributions of ViT (93.88%) and CNN (95.45%), demonstrating the potential for enhanced diagnostic accuracy in cardiovascular health monitoring. These results demonstrate the potential of ensemble methods in enhancing classification performance for cardiovascular health monitoring and diagnosis.

Problem

• Ensemble method combining CNN and ViT
• Improving heart sound classification accuracy
• Enhancing cardiovascular health monitoring

Innovation

• Ensemble of CNN and Transformer
• Mixture of Experts framework
• High heart sound classification accuracy