The Effect of the Loss on Generalization: Empirical Study on Synthetic Lung Nodule Data

Convolutional Neural Networks (CNNs) are widely used for image classification in a variety of fields, including medical imaging. While most studies deploy cross-entropy as the loss function in such tasks, a growing number of approaches have turned to a family of contrastive learning-based losses. Ev...

Full description

Saved in:
Bibliographic Details
Published in:arXiv.org 2021-08
Main Authors: Baltatzis, Vasileios, Loic Le Folgoc, Ellis, Sam, Martinez Manzanera, Octavio E, Kyriaki-Margarita Bintsi, Nair, Arjun, Desai, Sujal, Glocker, Ben, Schnabel, Julia A
Format: Article
Language:English
Subjects:
Artificial neural networks
Classifiers
Deep learning
Image classification
Lungs
Machine learning
Medical imaging
Performance measurement
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Convolutional Neural Networks (CNNs) are widely used for image classification in a variety of fields, including medical imaging. While most studies deploy cross-entropy as the loss function in such tasks, a growing number of approaches have turned to a family of contrastive learning-based losses. Even though performance metrics such as accuracy, sensitivity and specificity are regularly used for the evaluation of CNN classifiers, the features that these classifiers actually learn are rarely identified and their effect on the classification performance on out-of-distribution test samples is insufficiently explored. In this paper, motivated by the real-world task of lung nodule classification, we investigate the features that a CNN learns when trained and tested on different distributions of a synthetic dataset with controlled modes of variation. We show that different loss functions lead to different features being learned and consequently affect the generalization ability of the classifier on unseen data. This study provides some important insights into the design of deep learning solutions for medical imaging tasks.
ISSN:2331-8422