Convolutional Neural Networks for Histopathology Image Classification: Training vs. Using Pre-Trained Networks

TL;DR

This study compares the effectiveness of pre-trained CNN features versus training CNNs from scratch for histopathology image classification, highlighting the impact of transfer learning and fine-tuning on different network architectures.

Contribution

It provides an empirical evaluation of pre-trained versus trained-from-scratch CNNs on a large histopathology dataset, analyzing transfer learning effects and fine-tuning benefits.

Findings

Pre-trained networks perform competitively with trained-from-scratch models.

Fine-tuning VGG16 offers no significant improvement, unlike Inception.

Transfer learning is effective even with limited training samples.

Abstract

We explore the problem of classification within a medical image data-set based on a feature vector extracted from the deepest layer of pre-trained Convolution Neural Networks. We have used feature vectors from several pre-trained structures, including networks with/without transfer learning to evaluate the performance of pre-trained deep features versus CNNs which have been trained by that specific dataset as well as the impact of transfer learning with a small number of samples. All experiments are done on Kimia Path24 dataset which consists of 27,055 histopathology training patches in 24 tissue texture classes along with 1,325 test patches for evaluation. The result shows that pre-trained networks are quite competitive against training from scratch. As well, fine-tuning does not seem to add any tangible improvement for VGG16 to justify additional training while we observed considerable improvement in retrieval and classification accuracy when we fine-tuned the Inception structure.