Self-Supervised Learning for Building Robust Pediatric Chest X-ray Classification Models

TL;DR

This paper introduces SCC, a self-supervised learning approach that enhances pediatric chest X-ray classification by combining transfer learning, contrastive learning, and contrast enhancement, achieving superior performance across diverse datasets.

Contribution

The paper presents a novel SCC framework that improves pediatric CXR classification robustness and generalization using self-supervised contrastive learning combined with transfer learning and contrast enhancement techniques.

Findings

SCC outperforms transfer learning in zero-shot pediatric CXR classification by 13.6% and 34.6% AUC.

With 10x fewer labeled images, SCC matches full-data transfer learning performance.

SCC also improves breast cancer dataset classification, outperforming fully supervised models in zero-shot settings.

Abstract

Recent advancements in deep learning for Medical Artificial Intelligence have demonstrated that models can match the diagnostic performance of clinical experts in adult chest X-ray (CXR) interpretation. However, their application in the pediatric context remains limited due to the scarcity of large annotated pediatric image datasets. Additionally, significant challenges arise from the substantial variability in pediatric CXR images across different hospitals and the diverse age range of patients from 0 to 18 years. To address these challenges, we propose SCC, a novel approach that combines transfer learning with self-supervised contrastive learning, augmented by an unsupervised contrast enhancement technique. Transfer learning from a well-trained adult CXR model mitigates issues related to the scarcity of pediatric training data. Contrastive learning with contrast enhancement focuses on the lungs, reducing the impact of image variations and producing high-quality embeddings across diverse pediatric CXR images. We train SCC on one pediatric CXR dataset and evaluate its performance on two other pediatric datasets from different sources. Our results show that SCC's out-of-distribution (zero-shot) performance exceeds regular transfer learning in terms of AUC by 13.6% and 34.6% on the two test datasets. Moreover, with few-shot learning using 10 times fewer labeled images, SCC matches the performance of regular transfer learning trained on the entire labeled dataset. To test the generality of the framework, we verify its performance on three benchmark breast cancer datasets. Starting from a model trained on natural images and fine-tuned on one breast dataset, SCC outperforms the fully supervised learning baseline on the other two datasets in terms of AUC by 3.6% and 5.5% in zero-shot learning.