Transfer or Self-Supervised? Bridging the Performance Gap in Medical Imaging

TL;DR

This paper compares transfer learning and self-supervised learning in medical imaging, analyzing their performance, robustness, and suitability under various data challenges to guide effective application in medical research.

Contribution

It provides a comprehensive comparison of transfer and self-supervised learning methods in medical imaging, highlighting their advantages, limitations, and factors affecting their performance.

Findings

Transfer learning and self-supervised learning show different accuracy and robustness profiles.

Data imbalance, scarcity, and domain mismatch impact model performance.

Recommendations are provided for applying these methods effectively in medical contexts.

Abstract

Recently, transfer learning and self-supervised learning have gained significant attention within the medical field due to their ability to mitigate the challenges posed by limited data availability, improve model generalisation, and reduce computational expenses. Transfer learning and self-supervised learning hold immense potential for advancing medical research. However, it is crucial to recognise that transfer learning and self-supervised learning architectures exhibit distinct advantages and limitations, manifesting variations in accuracy, training speed, and robustness. This paper compares the performance and robustness of transfer learning and self-supervised learning in the medical field. Specifically, we pre-trained two models using the same source domain datasets with different pre-training methods and evaluated them on small-sized medical datasets to identify the factors influencing their final performance. We tested data with several common issues in medical domains, such as data imbalance, data scarcity, and domain mismatch, through comparison experiments to understand their impact on specific pre-trained models. Finally, we provide recommendations to help users apply transfer learning and self-supervised learning methods in medical areas, and build more convenient and efficient deployment strategies.