Cats or CAT scans: transfer learning from natural or medical image source datasets?

Abstract

Transfer learning is a widely used strategy in medical image analysis. Instead of only training a network with a limited amount of data from the target task of interest, we can first train the network with other, potentially larger source datasets, creating a more robust model. The source datasets do not have to be related to the target task. For a classification task in lung CT images, we could use both head CT images, or images of cats, as the source. While head CT images appear more similar to lung CT images, the number and diversity of cat images might lead to a better model overall. In this survey we review a number of papers that have performed similar comparisons. Although the answer to which strategy is best seems to be "it depends", we discuss a number of research directions we need to take as a community, to gain more understanding of this topic.

Accepted to Current Opinion in Biomedical Engineering

References in corpus (6)

• A Survey on Deep Learning in Medical Image Analysis
• Convolutional Neural Networks for Medical Image Analysis: Full Training or Fine Tuning?
• Why rankings of biomedical image analysis competitions should be interpreted with care
• Multi-source Transfer Learning with Convolutional Neural Networks for Lung Pattern Analysis
• Building medical image classifiers with very limited data using segmentation networks
• Exploring the similarity of medical imaging classification problems

Cited by in corpus (9)

• TorchIO: A Python library for efficient loading, preprocessing, augmentation and patch-based sampling of medical images in deep learning
• Checklist for responsible deep learning modeling of medical images based on COVID-19 detection studies
• Interpretable and intervenable ultrasonography-based machine learning models for pediatric appendicitis
• Exploring connections of spectral analysis and transfer learning in medical imaging
• 3D medical image segmentation with labeled and unlabeled data using autoencoders at the example of liver segmentation in CT images
• Task Fingerprinting for Meta Learning in Biomedical Image Analysis
• Automatic detection of rare pathologies in fundus photographs using few-shot learning
• Requirement analysis for an artificial intelligence model for the diagnosis of the COVID-19 from chest X-ray data
• Primary Tumor Origin Classification of Lung Nodules in Spectral CT using Transfer Learning