The FDS-Net deep learning model achieved a mean absolute error of 3.75 for predicting left ventricular ejection fraction on the EchoNet-Dynamic dataset, outperforming 11 benchmark methods.
Does FDS-Net improve LVEF prediction accuracy in echocardiograms compared to benchmark methods?
FDS-Net provides a robust deep learning solution for predicting LVEF in noisy echocardiograms by jointly modeling global spatiotemporal dependencies and fine-grained local motion.
Absolute Event Rate: 3.75% vs 3.93%
Abstract Left ventricular ejection fraction (LVEF) is a key clinical indicator of cardiac function, yet accurate prediction from echocardiography remains challenging due to speckle noise, anatomical ambiguity, and complex cardiac motion. We propose an optical Flow-enhanced Dual-Stream Network (FDS-Net) for LVEF prediction that jointly models global spatiotemporal dependencies and fine-grained local motion. FDS-Net consists of a UniFormer-based spatiotemporal stream to capture long-range cardiac-cycle context and a Mamba-based optical-flow stream to encode pixel-level left ventricular motion. An adaptive gate-driven fusion module dynamically integrates the two streams, enabling complementary use of structural-temporal semantics and motion cues. To improve the compatibility of optical flow with ultrasound, we further design a pretraining framework for the flow stream, including an Anatomy-Motion Semantic Fusion (AMSF) network that anchors motion learning to left ventricular anatomy via segmentation, and a Dynamically Regularized Enhanced Hybrid (DREH) loss that combines weakly supervised and unsupervised constraints. Experiments on EchoNet-Dynamic, Cardiac Acquisitions for Multi-structure Ultrasound Seg-mentation (CAMUS), and EchoHPPS show that FDS-Net achieves mean absolute error (MAE), root mean squared error (RMSE), and coefficient of determination (R2) values of 3.75/4.99/0.825, 5.58/7.56/0.680, and 4.13/5.92/0.711, respectively. Moreover, it outperforms 11 benchmark methods on all three datasets and provides a robust solution for predicting LVEF in noisy echocardiograms.
Deng et al. (Mon,) conducted a other in Left ventricular ejection fraction (LVEF) prediction (n=11,307). FDS-Net (Optical Flow-Enhanced Dual-Stream Network) vs. 11 benchmark deep learning methods (e.g., Echo-Vision-FM) was evaluated on Mean Absolute Error (MAE) for LVEF prediction on EchoNet-Dynamic dataset. The FDS-Net deep learning model achieved a mean absolute error of 3.75 for predicting left ventricular ejection fraction on the EchoNet-Dynamic dataset, outperforming 11 benchmark methods.