Privacy-Aware Continual Self-Supervised Learning on Multi-Window Chest Computed Tomography for Domain-Shift Robustness

TL;DR

This paper introduces a privacy-preserving continual self-supervised learning framework for chest CT images that effectively handles domain shifts caused by different window settings, improving robustness and generalization in medical diagnosis.

Contribution

The paper presents a novel CSSL method with latent replay and a combined WKD and BKE feature distillation technique to address domain shifts without compromising data privacy.

Findings

Outperforms existing methods in domain-shift robustness

Effectively mitigates catastrophic forgetting during continual pretraining

Enhances feature representations for medical image diagnosis

Abstract

We propose a novel continual self-supervised learning (CSSL) framework for simultaneously learning diverse features from multi-window-obtained chest computed tomography (CT) images and ensuring data privacy. Achieving a robust and highly generalizable model in medical image diagnosis is challenging, mainly because of issues, such as the scarcity of large-scale, accurately annotated datasets and domain shifts inherent to dynamic healthcare environments. Specifically, in chest CT, these domain shifts often arise from differences in window settings, which are optimized for distinct clinical purposes. Previous CSSL frameworks often mitigated domain shift by reusing past data, a typically impractical approach owing to privacy constraints. Our approach addresses these challenges by effectively capturing the relationship between previously learned knowledge and new information across different training stages through continual pretraining on unlabeled images. Specifically, by incorporating a latent replay-based mechanism into CSSL, our method mitigates catastrophic forgetting due to domain shifts during continual pretraining while ensuring data privacy. Additionally, we introduce a feature distillation technique that integrates Wasserstein distance-based knowledge distillation (WKD) and batch-knowledge ensemble (BKE), enhancing the ability of the model to learn meaningful, domain-shift-robust representations. Finally, we validate our approach using chest CT images obtained across two different window settings, demonstrating superior performance compared with other approaches.