Source Matters: Source Dataset Impact on Model Robustness in Medical Imaging

2024/03/07 by Dovile Juodelyte, Yucheng Lu, Juodelyte, Dovile +9 · 3 citations
Medicine · Computer Science · #Radiomics and Machine Learning in Medical Imaging #Medical Imaging Techniques and Applications #AI in cancer detection

paper · pdf · doi:10.48550/arxiv.2403.04484

Abstract

Transfer learning has become an essential part of medical imaging classification algorithms, often leveraging ImageNet weights. The domain shift from natural to medical images has prompted alternatives such as RadImageNet, often showing comparable classification performance. However, it remains unclear whether the performance gains from transfer learning stem from improved generalization or shortcut learning. To address this, we conceptualize confounders by introducing the Medical Imaging Contextualized Confounder Taxonomy (MICCAT) and investigate a range of confounders across it -- whether synthetic or sampled from the data -- using two public chest X-ray and CT datasets. We show that ImageNet and RadImageNet achieve comparable classification performance, yet ImageNet is much more prone to overfitting to confounders. We recommend that researchers using ImageNet-pretrained models reexamine their model robustness by conducting similar experiments. Our code and experiments are available at https://github.com/DovileDo/source-matters.

Cited by

• Revisiting Performance Claims for Chest X-Ray Models Using Clinical Context
• In-hoc Concept Representations to Regularise Deep Learning in Medical Imaging
• General Methods Make Great Domain-specific Foundation Models: A Case-study on Fetal Ultrasound

Related

• Revisiting Hidden Representations in Transfer Learning for Medical Imaging
• Transfusion: Understanding Transfer Learning for Medical Imaging
• MedNet: Pre-trained Convolutional Neural Network Model for the Medical Imaging Tasks
• Performance Evaluation of Transfer Learning Based Medical Image Classification Techniques for Disease Detection
• Disease Classification and Impact of Pretrained Deep Convolution Neural Networks on Diverse Medical Imaging Datasets across Imaging Modalities