Leveraging Self-Supervised Learning for Fetal Cardiac Planes Classification using Ultrasound Scan Videos

TL;DR

This paper explores the use of self-supervised learning on ultrasound videos to improve fetal heart plane classification, demonstrating that dataset variance is key and BarlowTwins offers robust transfer learning performance with limited labeled data.

Contribution

It introduces a comprehensive evaluation of seven SSL methods for fetal ultrasound video analysis and highlights the effectiveness of BarlowTwins for transfer learning in this domain.

Findings

Dataset variance impacts SSL effectiveness more than dataset size.

BarlowTwins provides consistent performance across different settings.

Fine-tuning with 1% labeled data surpasses ImageNet initialization by 12% in F1-score.

Abstract

Self-supervised learning (SSL) methods are popular since they can address situations with limited annotated data by directly utilising the underlying data distribution. However, the adoption of such methods is not explored enough in ultrasound (US) imaging, especially for fetal assessment. We investigate the potential of dual-encoder SSL in utilizing unlabelled US video data to improve the performance of challenging downstream Standard Fetal Cardiac Planes (SFCP) classification using limited labelled 2D US images. We study 7 SSL approaches based on reconstruction, contrastive loss, distillation, and information theory and evaluate them extensively on a large private US dataset. Our observations and findings are consolidated from more than 500 downstream training experiments under different settings. Our primary observation shows that for SSL training, the variance of the dataset is more crucial than its size because it allows the model to learn generalisable representations, which improve the performance of downstream tasks. Overall, the BarlowTwins method shows robust performance, irrespective of the training settings and data variations, when used as an initialisation for downstream tasks. Notably, full fine-tuning with 1% of labelled data outperforms ImageNet initialisation by 12% in F1-score and outperforms other SSL initialisations by at least 4% in F1-score, thus making it a promising candidate for transfer learning from US video to image data.