AI Black Box
Published in

AI Black Box

[Review]Transfusion: Understanding Transfer Learning for Medical Imaging


Recognition and appreciation

For your notice

What is transfer learning?

  • Performance
  • Learned representations and features
  • Weight scalings

Deep learning trend

Deep learning medical imaging
Machine learning medical imaging
  • COCO dataset experiment: This experiment carried out by Kaiming et al., in Rethinking ImageNet Pre-training whereby instance segmentation and object detection were performed using models trained from random initialisation and ImageNet pretrained models. The results from both models obtained were not far apart which brings the performance ability of transfer learning in question.
  • In Do better ImageNet models transfer better by Simon et al.,experiment result shows that ImageNet architectures such as VGG16 generalizes across datasets but the features are less general regardless of the dimensionality reduction strategy employed. This is because ImageNet features and medical images features have considerable differences.
Sample images from ImageNet: Images clearly and explicitly depicts a universal class
Retina Fundus Photographs: Small red ‘dots’ are an indication of microaneurysms and diabetic retinopathy
Sample images from CheXpert: Chest x-rays local white opaque patches are signs of consolidation and pneumonia.

Paper Contributions

Transfer learning and random initialization perform comparably across both standard ImageNet architectures and simple, lightweight CNNs from RETINA dataset for diagnosing moderate Diabetes Retinopathy.
Transfer learning provides mixed performance gains on chest x-rays. Performances (AUC%) of diagnosing different pathologies on the CheXpert dataset

Further readings

  1. A Gentle Introduction to Transfer Learning for Deep Learning
  2. Transfer Learning



Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store
Tobiloba Adejumo

Interested in biomarker development, software dev and ai, as well as psychology, history, philosophy, relationships. Website: