Beyond Doppler: Scalable AI Detection of LVOT Obstruction in HCM.

Dual-flow convolutional neural network for automatic measurement of left ventricular ejection fraction and global longitudinal strain in echocardiography

Echo-DFCNN achieves simultaneous and precise assessment of LVEF and GLS in the study cohorts, demonstrating its potential for robust performance across a wide range of cardiac functions, different image qualities, and machine types.

Automated Detection of Clinically Significant Mitral Regurgitation from Single-View B-mode Echocardiography Using Deep Learning.

MitralVision reliably distinguishes clinically significant MR using single-view B-mode echocardiography without Doppler input for model inference and may support more standardized MR screening.

Learning from Scarce Labels: Multi-View Echocardiography for Ejection Fraction Prediction

We present, to the best of our knowledge, the first publicly available resource for predicting left ventricular ejection fraction (EF) from parasternal long-axis (PLAX) echocardiography. Because no PLAX–EF datasets previously existed, our work focuses on an innovative data generation strategy to overcome this scarcity. By leveraging a time-based correlation between clinical notes and echocardiographic videos, combined with fine-tuning view classifiers and proxy labeling, we created a labeled dataset of over 25,000 PLAX videos. This enables us to train the first reproducible PLAX EF model, achieving a mean absolute error (MAE) of 6.86%. Given that apical four-chamber (A4C) methods, the clinical standard, report MAE values of 6%-7%, our results demonstrate that EF estimation from PLAX views is both feasible and clinically relevant. This surpasses the performance of existing methods and provides a clinically relevant solution for situations where apical views may not be feasible. Going further, we demonstrate that combining PLAX and A4C predictions via simple unweighted late fusion improves both single-view baselines to a 6.37% MAE, underscoring the value of multi-view integration. To promote continued research, we release the dataset labels, trained models, and runnable demos on GitHub, Hugging Face, and Google Colab: https://github.com/Jeffrey4899/PLAX_EF_Labels_202509

AI-PPG for pre-screening of reduced left ventricular ejection fraction: development and prospective external validation

ESPNet: An Automated Left Ventricular Ejection Fraction Measurement via Multi-Frame Temporal Aggregation.

Left ventricular ejection fraction (LVEF) is a key indicator of cardiac function assessed via echocardiography. The biplane Simpson's method for manual measurement is time-consuming and subjective, requiring identification of apical four-chamber (A4C) and two-chamber (A2C) views at end-diastolic (ED) and end-systolic (ES) frames. Automated methods often struggle with consistency and reliability, particularly in heart failure, arrhythmia and low signal-to-noise ratio (SNR) conditions. To address these challenges, this study presents EchoSegPhaseNet (ESP Net), an innovative automated approach for LVEF measurement using multi-frame temporal aggregation. ESP Net comprises two main components: the PhaseFormer module for cardiac phase detection and the LVFormer for Endo segmentation. PhaseFormer significantly improves ED/ES frame prediction accuracy, primarily due to the Multi Probability Prediction Head (MPP), which enables long range temporal receptive fields and effectively aggregates deep-layer features over various distances in the echocar diographic video stream. Following this, LVFormer aggregates deep temporal features from frames around ED/ES, combining spatial-temporal features from one complete respective cardiac cycle of both A2C and A4C views to overcome single-view limitations and significantly enhance segmentation accuracy. This upstream-downstream pro cessing pipeline significantly boosts data processing efficiency. In evaluations on both private and public datasets, ESPNet showcases a faster inference speed in phase detection compared to existing methods, while maintaining similar accuracy levels. In the Endo segmentation task, it not only achieves superior speed but also outperforms current methods in accuracy, with a Pearson correlation coefficient of 0.981, an R2 of 0.96, and a mean absolute error of 3.281 between automated LVEF measurements and manual physician calculations. These results demonstrate that ESPNet not only outperforms existing methods but also achieves a level of consistency comparable to inter observer variability among mid-career clinicians, effectively addressing the clinical need for reliable and efficient automated LVEF measurement.

ROI-Guided Echocardiographic Image Analysis and Case Aggregation for Differentiating Atrial and Ventricular Septal Defects

Echocardiography is the primary imaging modality for congenital heart disease (CHD) assessment. However, the real-world clinical application of artificial intelligence in this domain is often hindered by hidden data leakage from homologous frames and the high diagnostic variance of single-frame static inference. To bridge the gap between idealized model evaluation and clinical reality, this study proposes a standardized, region-of-interest (ROI)-guided deep learning workflow for differentiating atrial septal defect (ASD), ventricular septal defect (VSD), and normal cases. Specifically, utilizing a retrospective cohort of 1,987 patients (11,638 images) from Tengzhou Central People’s Hospital, we implemented strict subject-disjoint partitioning to eliminate data leakage, and simultaneously introduced cross-frame case aggregation to emulate the multi-frame visual synthesis process of expert echocardiographers. Among the evaluated model configurations, fully fine-tuned EfficientNet-B0 achieved the highest internal held-out test performance, with an accuracy of 0.9950 (95% CI, 0.9875–1.0000), a macro-F1 of 0.9941, and a macro-AUROC of 0.9997; 2 of 401 test patients were misclassified. Five-fold patient-level cross-validation yielded an accuracy of $0.9945 \pm 0.0040$ , further supporting the stability of the model across patient partitions. These findings suggest that the proposed workflow has the potential to serve as an adjunctive tool for septal defect screening.