In this section
@article{Naidoo:2025:10.1016/j.compbiomed.2025.111148,
author = {Naidoo, P and Fernandes, P and Dadashi, Serej N and Manisty, CH and Shun-Shin, MJ and Francis, DP and Zolgharni, M},
doi = {10.1016/j.compbiomed.2025.111148},
journal = {Comput Biol Med},
title = {Consensus-guided evaluation of self-supervised learning in echocardiographic segmentation.},
url = {http://dx.doi.org/10.1016/j.compbiomed.2025.111148},
volume = {198},
year = {2025}
}
TY - JOUR
AB - BACKGROUND: 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. METHODS: 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. RESULTS: 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. CONCLUSION: The findings suggest that AI models can generalise well across expert annotations, providing more reliable and reproducible assessments.
AU - Naidoo,P
AU - Fernandes,P
AU - Dadashi,Serej N
AU - Manisty,CH
AU - Shun-Shin,MJ
AU - Francis,DP
AU - Zolgharni,M
DO - 10.1016/j.compbiomed.2025.111148
PY - 2025///
TI - Consensus-guided evaluation of self-supervised learning in echocardiographic segmentation.
T2 - Comput Biol Med
UR - http://dx.doi.org/10.1016/j.compbiomed.2025.111148
UR - https://www.ncbi.nlm.nih.gov/pubmed/41038129
VL - 198
ER -