# Convolutional Neural Networks for Medical Image Analysis: Full Training or Fine Tuning?

## TL;DR

This paper investigates whether fine-tuning pre-trained CNNs can replace training from scratch in medical image analysis, showing that fine-tuning often yields equal or better performance and is more robust to data size.

## Contribution

The study systematically compares full training and layer-wise fine-tuning of CNNs across multiple medical imaging tasks, providing practical guidelines for optimal transfer learning strategies.

## Key findings

- Pre-trained CNNs with fine-tuning outperform or match training from scratch.
- Fine-tuned CNNs are more robust to training data size.
- Layer-wise fine-tuning helps optimize performance based on data availability.

## Abstract

Training a deep convolutional neural network (CNN) from scratch is difficult because it requires a large amount of labeled training data and a great deal of expertise to ensure proper convergence. A promising alternative is to fine-tune a CNN that has been pre-trained using, for instance, a large set of labeled natural images. However, the substantial differences between natural and medical images may advise against such knowledge transfer. In this paper, we seek to answer the following central question in the context of medical image analysis: \emph{Can the use of pre-trained deep CNNs with sufficient fine-tuning eliminate the need for training a deep CNN from scratch?} To address this question, we considered 4 distinct medical imaging applications in 3 specialties (radiology, cardiology, and gastroenterology) involving classification, detection, and segmentation from 3 different imaging modalities, and investigated how the performance of deep CNNs trained from scratch compared with the pre-trained CNNs fine-tuned in a layer-wise manner. Our experiments consistently demonstrated that (1) the use of a pre-trained CNN with adequate fine-tuning outperformed or, in the worst case, performed as well as a CNN trained from scratch; (2) fine-tuned CNNs were more robust to the size of training sets than CNNs trained from scratch; (3) neither shallow tuning nor deep tuning was the optimal choice for a particular application; and (4) our layer-wise fine-tuning scheme could offer a practical way to reach the best performance for the application at hand based on the amount of available data.