A fully automated machine learning assisted pipeline to predict disease progression for early-stage hypertrophic cardiomyopathy from echocardiography

BackMix-Enhanced Semi-Supervised Learning for Automated Detection of Aortic Stenosis from Transthoracic Echocardiographic Images

Aortic stenosis (AS) is a prevalent and progressive valvular heart disease requiring accurate severity assessment for optimal clinical decision-making. Transthoracic echocardiography (TTE) is the standard diagnostic modality; however, its interpretation remains operator-dependent and subject to inter-observer variability. In this study, we propose an anatomically guided BackMix-enhanced semi-supervised learning framework for automated echocardiographic view classification and AS severity assessment. The approach leverages Gradient-weighted Class Activation Mapping (Grad-CAM) to preserve diagnostically relevant anatomical regions during augmentation while modifying background areas to mitigate shortcut learning. A semi-supervised self-training strategy combined with an ensemble classification framework was used to exploit both labeled and unlabeled data. The framework was evaluated on the TMED2 dataset of echocardiography, comprising 24,964 TTE images. To assess generalizability, external validation was performed on an independent dataset of 300 cardiologist-labeled echocardiographic images, including apical four-chamber (A4C), parasternal long-axis (PLAX), and parasternal short-axis (PSAX) views. Experimental results demonstrated that the proposed method outperformed the baseline semi-supervised model, improving view classification accuracy by approximately 3% and AS severity classification accuracy by 4–6%, with the largest gain observed in moderate AS. Performance remained consistent on the external validation dataset, supporting the robustness of the proposed approach. Statistical analysis confirmed the significance of these improvements (p < 0.01). Grad-CAM evaluation further demonstrated improved localization of clinically relevant regions. These findings suggest that anatomically guided BackMix augmentation combined with semi-supervised ensemble learning can improve classification accuracy, robustness, and interpretability in echocardiographic analysis under limited annotation conditions, offering a promising approach for automated AS assessment across independent clinical datasets

Artificial intelligence-empowered echocardiography: an updated review of clinical management in hypertrophic cardiomyopathy

Hypertrophic cardiomyopathy (HCM) is a common and highly heterogeneous inherited cardiomyopathy characterized by complex clinical phenotypes and diverse disease trajectories, posing significant challenges for early diagnosis, precise phenotypic classification, and risk stratification. Owing to its noninvasive nature, repeatability, and wide availability, echocardiography remains the cornerstone imaging modality for the diagnosis and longitudinal management of HCM. However, conventional echocardiographic analysis relies heavily on operator expertise and is limited in its ability to comprehensively extract latent structural, functional, and tissue-level information embedded within imaging data. In recent years, artificial intelligence (AI), particularly deep learning, has undergone rapid development in automated echocardiographic analysis, enabling a paradigm shift from traditional morphology-based assessment toward data-driven intelligent decision-support platforms. This review systematically categorizes AI-based methods in echocardiography according to the complexity of data processing, ranging from single-frame structural and texture analysis to spatiotemporal modeling of cardiac function, multi-view representation learning, and ultimately multimodal integration incorporating diverse clinical data sources. We further summarize the clinical applications of these AI-based methods in HCM, including diagnosis and differential diagnosis, phenotype characterization, and risk prediction. In addition, current challenges are discussed, including limited interpretability, data heterogeneity, and insufficient large-scale clinical validation, and future research directions are proposed. Overall, AI-empowered echocardiography holds substantial promise for advancing precision diagnosis, risk stratification, and personalized management of HCM, facilitating a transition toward more intelligent and individualized cardiovascular care.

External validation of machine learning models predicting global longitudinal strain from conventional echocardiography in patients with cancer.

Validation of a deep learning-based workflow for the interpretation of the echocardiogram in a cardiooncology population

Echocardiography is the cornerstone for risk stratification, diagnosis, and monitoring of cancer therapy–related cardiac dysfunction (CTRCD)(1). Artificial intelligence (AI)–guided echocardiography has shown high accuracy and reliability in diverse cardiac populations and may reduce variability while improving workflow efficiency(2, 3). However, this technology has not yet been validated in a dedicated cohort of patients with cancer. To evaluate the accuracy and reliability of AI-guided echocardiography in assessing left ventricular ejection fraction (LVEF) and additional parameters, compared with conventional echocardiography, in a cardio-oncology population. This study included patients identified retrospectively from a cardio-oncology registry. Studies, that had already been analysed manually by expert sonographers and reported using AGFA PACS system, were uploaded to the US2.ai platform for automated analysis. The primary outcome was the level of agreement (LoA) between AI-guided and standard echocardiography for LVEF. Secondary outcomes included LoA for additional echocardiographic parameters and LoA between AI- LVEF and 3D LVEF. The performance of the deep learning (DL) algorithm in identifying LVEF <50% was evaluated using the area under the receiver operating characteristic curve (ROC-AUC). Subgroup analyses were performed in predefined populations clinically relevant in cardio-oncology. A total of 282 patients were included. Mean age was 60 ± 16 years, and 61% were women. Breast cancer was the most frequent malignancy (30.5%), followed by haematological malignancies (16.7%) and gastrointestinal tumours (10.6%). Manual median 2D LVEF was 60% (IQR: 55-64) and AI-derived LVEF was 59.2% (IQR: 53-64) showing good agreement and correlation (bias: −0.138, SD: 5.38, 95% LoA: −10.7 to 10.4, ICC: 0.791, 95% CI: 0.742–0.831, Spearman ρ: 0.718,), Table 1. Comparison between 3D echocardiography LVEF and AI-derived 2D LVEF showed similar agreement with narrower limits (bias: −0.13, 95% LoA: −9.51 to 9.26). The DL algorithm accurately identified LVEF <50% (ROC-AUC: 0.918, 95% CI: 0.875–0.961), Figure 1. Subgroup analyses demonstrated consistent agreement in patients with breast cancer, body mass index >30, prior radiotherapy and pericardial effusion. In a large real-world cardio-oncology cohort, AI-guided echocardiography demonstrated strong agreement with conventional echocardiography for LVEF assessment and high accuracy for detecting clinically relevant LV dysfunction. Performance was consistent across key subgroups, supporting the feasibility, reliability, and potential clinical value of integrating DL-based analysis into routine cardio-oncology echocardiographic workflows.Agreement between manual and AI-echo  AUC-ROC curve for LVEF<50%

Machine learning algorithms for predicting arrhythmic events in Hypertrophic Cardiomyopathy: limited enhancement beyond late gadolinium enhancement.

Deep Learning-Based Classification of Apical Hypertrophic Cardiomyopathy Using Left Ventricular Opacification and Enhanced LV Apex Imaging.

INTRODUCTION Accurate classification of apical hypertrophic cardiomyopathy (ApHCM) subtypes is challenging due to morphological variability and overlapping phenotypes. Conventional echocardiography provides limited visualization of the apex. Artifacts induced during left ventricular opacification (LVO) complicate diagnostic interpretation. A deep learning-based framework enhances image quality and improves subtype classification. MATERIALS AND METHODS In this work, a deep learning-based framework is used for ApHCM subtype classification. Apical four-chamber end-diastolic frames from 3,200 individual patients were extracted from the EchoNet-Dynamic Dataset. Two cardiology experts manually annotated images into pure ApHCM, relative ApHCM, mixed ApHCM, and normal classes, based on apical wall thickness and morphological characteristics, using a computer vision annotation tool. A deep learning pipeline integrated multilevel graph-based adaptive particle swarm optimization with a deep denoised convolutional neural network (MG-APSO-DnCNN) to suppress reverberation and clutter artifacts from LVO echocardiograms. Enhanced images were then segmented using a U-Net-based levelset model to delineate the left ventricular (LV) apex. Morphological and LV wall features were extracted from the segmented region, and a graph isomorphism network (GIN) was trained to capture both local hypertrophic patterns and global ventricular morphology for subtype classification. The framework was designed to distinguish among pure ApHCM, relative ApHCM, and mixed ApHCM. RESULTS The framework achieved a classification accuracy of 96.2%, with a precision, recall, and F1-score of approximately 95%. Cross-validation results indicate stable performance (95.8% ± 0.4%, p < 0.001), and ablation experiments confirmed the contribution of each pipeline component. DISCUSSION By combining denoising, segmentation, and graph-based learning, the framework addressed limitations caused by LVO artifacts and improved the recognition of subtle ApHCM subtypes. These results demonstrate the clinical potential of integrating morphological feature extraction with deep learning in echocardiography. CONCLUSION The framework integrating MG-APSO-DnCNN and GIN enables accurate and robust ApHCM subtype classification, supporting cardiologists in early diagnosis, patient risk stratification, and treatment planning.