Echo-E$^3$Net: Efficient Endo-Epi Spatio-Temporal Network for Ejection Fraction Estimation

2025-03-21

AI Summary

Traditional LVEF assessment methods are time-consuming and operator-dependent, while existing deep learning models incur high computational costs and neglect spatiotemporal协同 modeling of endocardial and epicardial boundaries. To address these limitations, this paper proposes E²Net—a lightweight, end-to-end convolutional neural network. E²Net introduces two novel components: the Endo-Epi Cardial Border Detector (E²CBD) and the Feature Aggregator (E²FA), jointly capturing dynamic spatiotemporal boundary evolution and statistical features of both myocardial layers. It employs a multi-component loss function explicitly aligned with the clinical definition of ejection fraction and adopts a pretraining-free, data-augmentation-free, single-model lightweight CNN backbone integrated with spatiotemporal alignment and boundary-guided attention mechanisms. Evaluated on EchoNet-Dynamic, E²Net achieves RMSE = 5.15 and R² = 0.82, with only 6.8M parameters and 8.49G FLOPs—enabling real-time deployment in point-of-care ultrasound (PoCUS) settings.

Technology Category

• Machine Learning: Learning on the Edge & Model Compression
• Computer Vision: Multi-modal Vision
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Energy management for devices in mobile Web and WoT environments
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Left ventricular ejection fraction (LVEF) is a critical metric for assessing cardiac function, widely used in diagnosing heart failure and guiding clinical decisions. Despite its importance, conventional LVEF estimation remains time-consuming and operator-dependent. Recent deep learning advancements have enhanced automation, yet many existing models are computationally demanding, hindering their feasibility for real-time clinical applications. Additionally, the interplay between spatial and temporal features is crucial for accurate estimation but is often overlooked. In this work, we propose Echo-E$^3$Net, an efficient Endo-Epi spatio-temporal network tailored for LVEF estimation. Our method introduces the Endo-Epi Cardial Border Detector (E$^2$CBD) module, which enhances feature extraction by leveraging spatial and temporal landmark cues. Complementing this, the Endo-Epi Feature Aggregator (E$^2$FA) distills statistical descriptors from backbone feature maps, refining the final EF prediction. These modules, along with a multi-component loss function tailored to align with the clinical definition of EF, collectively enhance spatial-temporal representation learning, ensuring robust and efficient EF estimation. We evaluate Echo-E$^3$Net on the EchoNet-Dynamic dataset, achieving a RMSE of 5.15 and an R$^2$ score of 0.82, setting a new benchmark in efficiency with 6.8 million parameters and only 8.49G Flops. Our model operates without pre-training, data augmentation, or ensemble methods, making it well-suited for real-time point-of-care ultrasound (PoCUS) applications. Our Code is publicly available on~href{https://github.com/moeinheidari7829/Echo-E3Net}{ extcolor{magenta}{GitHub}}.

Problem

• Automates LVEF estimation for cardiac function assessment
• Reduces computational demand for real-time clinical use
• Improves spatial-temporal feature integration in EF prediction

Innovation

• Endo-Epi Cardial Border Detector enhances feature extraction
• Endo-Epi Feature Aggregator distills statistical descriptors
• Multi-component loss aligns with clinical EF definition