MedAug: Contrastive learning leveraging patient metadata improves representations for chest X-ray interpretation

Abstract

Self-supervised contrastive learning between pairs of multiple views of the same image has been shown to successfully leverage unlabeled data to produce meaningful visual representations for both natural and medical images. However, there has been limited work on determining how to select pairs for medical images, where availability of patient metadata can be leveraged to improve representations. In this work, we develop a method to select positive pairs coming from views of possibly different images through the use of patient metadata. We compare strategies for selecting positive pairs for chest X-ray interpretation including requiring them to be from the same patient, imaging study or laterality. We evaluate downstream task performance by fine-tuning the linear layer on 1% of the labeled dataset for pleural effusion classification. Our best performing positive pair selection strategy, which involves using images from the same patient from the same study across all lateralities, achieves a performance increase of 14.4% in mean AUC from the ImageNet pretrained baseline. Our controlled experiments show that the keys to improving downstream performance on disease classification are (1) using patient metadata to appropriately create positive pairs from different images with the same underlying pathologies, and (2) maximizing the number of different images used in query pairing. In addition, we explore leveraging patient metadata to select hard negative pairs for contrastive learning, but do not find improvement over baselines that do not use metadata. Our method is broadly applicable to medical image interpretation and allows flexibility for incorporating medical insights in choosing pairs for contrastive learning.

References in corpus (11)

• Improved Baselines with Momentum Contrastive Learning
• Learning deep representations by mutual information estimation and maximization
• What Makes for Good Views for Contrastive Learning?
• Big Self-Supervised Models are Strong Semi-Supervised Learners
• Contrastive Learning of Medical Visual Representations from Paired Images and Text
• Unsupervised Feature Learning via Non-Parametric Instance-level Discrimination
• MoCo-CXR: MoCo Pretraining Improves Representation and Transferability of Chest X-ray Models
• COVID-19 Prognosis via Self-Supervised Representation Learning and Multi-Image Prediction
• CLOCS: Contrastive Learning of Cardiac Signals Across Space, Time, and Patients
• Viewmaker Networks: Learning Views for Unsupervised Representation Learning
• Deep Learning for the Digital Pathologic Diagnosis of Cholangiocarcinoma and Hepatocellular Carcinoma: Evaluating the Impact of a Web-based Diagnostic Assistant

Cited by in corpus (8)

• Recent advances and clinical applications of deep learning in medical image analysis
• Self-supervised learning methods and applications in medical imaging analysis: A survey
• 3KG: Contrastive Learning of 12-Lead Electrocardiograms using Physiologically-Inspired Augmentations
• MedSelect: Selective Labeling for Medical Image Classification Combining Meta-Learning with Deep Reinforcement Learning
• Imbalance-Aware Self-Supervised Learning for 3D Radiomic Representations
• USCL: Pretraining Deep Ultrasound Image Diagnosis Model through Video Contrastive Representation Learning
• SCALP -- Supervised Contrastive Learning for Cardiopulmonary Disease Classification and Localization in Chest X-rays using Patient Metadata
• Contrastive Learning with Temporal Correlated Medical Images: A Case Study using Lung Segmentation in Chest X-Rays