Transfusion: Understanding Transfer Learning for Medical Imaging
Authors:
Preprint
Abstract
Transfer learning from natural image datasets, particularly ImageNet, using standard large models and corresponding pretrained weights has become a de-facto method for deep learning applications to medical imaging. However, there are fundamental differences in data sizes, features and task specifications between natural image classification and the target medical tasks, and there is little understanding of the effects of transfer. In this paper, we explore properties of transfer learning for medical imaging. A performance evaluation on two large scale medical imaging tasks shows that surprisingly, transfer offers little benefit to performance, and simple, lightweight models can perform comparably to ImageNet architectures. Investigating the learned representations and features, we find that some of the differences from transfer learning are due to the over-parametrization of standard models rather than sophisticated feature reuse. We isolate where useful feature reuse occurs, and outline the implications for more efficient model exploration. We also explore feature independent benefits of transfer arising from weight scalings.Keywords
All Related Versions
This publication has 11 references indexed in Scilit:
- Rethinking ImageNet Pre-Training2019
- Efficient Deep Network Architectures for Fast Chest X-Ray Tuberculosis Screening and VisualizationScientific Reports, 2019
- A Deep Learning Model to Predict a Diagnosis of Alzheimer Disease by Using 18F-FDG PET of the BrainRadiology, 2019
- High-performance medicine: the convergence of human and artificial intelligenceNature Medicine, 2019
- Clinically applicable deep learning for diagnosis and referral in retinal diseaseNature Medicine, 2018
- Validation of automated screening for referable diabetic retinopathy with the IDx‐DR device in the Hoorn Diabetes Care SystemActa Ophthalmologica, 2017
- Dermatologist-level classification of skin cancer with deep neural networksNature, 2017
- Development and Validation of a Deep Learning Algorithm for Detection of Diabetic Retinopathy in Retinal Fundus PhotographsJAMA, 2016
- Improved Automated Detection of Diabetic Retinopathy on a Publicly Available Dataset Through Integration of Deep LearningInvestigative Ophthalmology & Visual Science, 2016
- Diabetic retinopathy – Biomolecules and multiple pathophysiologyDiabetes & Metabolic Syndrome: Clinical Research & Reviews, 2015