7 March 2024 in Computer Methods and Programs in Biomedicine
Training deep learning models for medical image segmentation require large annotated datasets, which can be expensive and time-consuming to create. Active learning is a promising approach to reduce this burden by strategically selecting the most informative samples for segmentation. This study investigates the use of active learning for efficient left ventricle segmentation in echocardiography with sparse expert annotations. We adapt and evaluate various sampling techniques, demonstrating their effectiveness in judiciously selecting samples for segmentation. Additionally, we introduce a novel strategy, Optimised Representativeness Sampling, which combines feature-based outliers with the most representative samples to enhance annotation efficiency. Our findings demonstrate a substantial reduction in annotation costs, achieving a remarkable 99% upper bound performance while utilising only 20% of the labelled data. This equates to a reduction of 1680 images needing annotation within our dataset. When applied to a publicly available dataset, our approach yielded a remarkable 70% reduction in required annotation efforts, representing a significant advancement compared to baseline active learning strategies, which achieved only a 50% reduction. Our experiments highlight the nuanced performance of diverse sampling strategies across datasets within the same domain. The study provides a cost-effective approach to tackle the challenges of limited expert annotations in echocardiography. By introducing a distinct dataset, made publicly available for research purposes, our work contributes to the field's understanding of efficient annotation strategies in medical image segmentation.
27 February 2026 in Medical Image Analysis
Accurate classification of echocardiographic views is fundamental for automated cardiac analysis. However, clinical practice relies on a large, heterogeneous set of fine-grained acquisitions that introduce substantial inter-observer variability. Existing studies have primarily focused on limited view sets, often collapsing specialised views into broad categories, which limits their clinical relevance. We address this limitation by introducing TTE47, the first publicly available benchmark comprising 47 clinically meaningful views annotated independently by three experts. This dataset enables the rigorous quantification of inter-observer agreement and establishes a foundation for reproducible, clinically relevant evaluation. To tackle the dual challenges of subtle inter-class distinctions and structured label variability, we propose a novel supervised contrastive learning framework incorporating a tailored loss function. Our method outperforms cross-entropy and standard supervised contrastive baselines, achieving leading performance among evaluated methods on TTE47 and surpassing prior work on TMED-2 without dataset-specific pretraining, using a model pretrained on TTE47. Beyond accuracy, we introduce clustering-based metrics, Detection Rate and Label Recovery Precision, that measure semantic coherence and the model's ability to resist annotation variability. Results show that the learned feature space aligns more strongly with underlying anatomical structure than with any single annotator's style, enabling resilience to label shifts and maintaining robustness comparable to human-level disagreement. By integrating multi-expert evaluation, robust representation learning, and interpretable feature-space analysis, this work establishes a scalable and clinically relevant framework for fine-grained echo view classification. Our findings highlight the potential of contrastive pretraining to standardise interpretation, mitigate subjectivity, and enhance the reliability of AI-assisted echocardiography in diverse clinical settings.
1 October 2025 in Computers in Biology and Medicine
Left ventricle segmentation is a fundamental task in echocardiography, essential for assessing cardiac function. However, deep learning models for segmentation rely on large labelled datasets, which are expensive and time-consuming to annotate. Self-supervised learning has emerged as a promising approach to leverage unlabelled data, but its effectiveness for left ventricle segmentation remains underexplored. This study investigates self-supervised learning for echocardiographic segmentation, comparing various pretext tasks. The impact of dataset size and distribution on pre-training is examined, revealing that excessive unlabelled data can degrade performance due to redundancy and low variability. A novel multi-expert labelled dataset is introduced to enhance segmentation evaluation, using consensus-based annotations to reduce annotation noise and improve reliability. Among the self-supervised learning methods evaluated, contrastive learning consistently outperforms other approaches, particularly in low-label settings. The study demonstrates that AI models pre-trained using self-supervised learning and fine-tuned with only 15% of labelled data achieve stronger alignment with multi-expert consensus than any individual expert. The findings suggest that AI models can generalise well across expert annotations, providing more reliable and reproducible assessments.
23 February 2026 in SSRN Electronic Journal
Background and ObjectiveMedical image segmentation is crucial in diagnosing and treating various diseases, often requiring extensive manual annotation for training deep learning models. This study explores the efficiency of semisupervised active learning for left ventricle segmentation in echocardiography, aiming to alleviate the need for large annotated datasets. A new technique for identifying reliable pseudo-labels is proposed, aiming to address the challenge of sparse expert annotations.MethodsWe proposed the resource-optimised semi-supervised active learning approach that selects reliable pseudo-labels from a range of samples with the highest uncertainty frequency based on their post-sorting confidence scores. Moreover, we incorporated a post-processing step for the predicted masks, refining them before their adoption as pseudo-labels. We compared our proposed method to the state-of-the-art active learning methods to show its efficiency. ResultsOur results reveal a significant decrease in annotation expenses, achieving a remarkable performance of 99% upper bound while utilising only 7% of the labelled data. Our approach achieved a 17% reduction in annotation efforts compared to competing active learning methods, marking a significant advancement over baseline strategies that required 24% annotation.ConclusionsThe study presents a resource-optimised approach to address the concern of limited expert annotations in echocardiography. By introducing a unique dataset, made publicly accessible for research endeavours, our work contributes to the field of efficient annotation strategies in medical image segmentation. Future investigations should prioritise extending the applicability of our proposed technique to other medical imaging domains and refining its algorithm for broader use.