<i>DreamOn:</i> a data augmentation strategy to narrow the robustness gap between expert radiologists and deep learning classifiers

There is no author summary for this article yet. Authors can add summaries to their articles on ScienceOpen to make them more accessible to a non-specialist audience.

Abstract

Purpose

Successful performance of deep learning models for medical image analysis is highly dependent on the quality of the images being analysed. Factors like differences in imaging equipment and calibration, as well as patient-specific factors such as movements or biological variability (e.g., tissue density), lead to a large variability in the quality of obtained medical images. Consequently, robustness against the presence of noise is a crucial factor for the application of deep learning models in clinical contexts.

Materials and methods

We evaluate the effect of various data augmentation strategies on the robustness of a ResNet-18 trained to classify breast ultrasound images and benchmark the performance against trained human radiologists. Additionally, we introduce DreamOn, a novel, biologically inspired data augmentation strategy for medical image analysis. DreamOn is based on a conditional generative adversarial network (GAN) to generate REM-dream-inspired interpolations of training images.

Results

We find that while available data augmentation approaches substantially improve robustness compared to models trained without any data augmentation, radiologists outperform models on noisy images. Using DreamOn data augmentation, we obtain a substantial improvement in robustness in the high noise regime.

Conclusions

We show that REM-dream-inspired conditional GAN-based data augmentation is a promising approach to improving deep learning model robustness against noise perturbations in medical imaging. Additionally, we highlight a gap in robustness between deep learning models and human experts, emphasizing the imperative for ongoing developments in AI to match human diagnostic expertise.

Related collections

Most cited references 43

Record: found
Abstract: not found
Article: not found

The Measurement of Observer Agreement for Categorical Data

J Landis, Gary G. Koch (1977)

0 comments Cited 8410 times – based on 0 reviews      Review now

Bookmark

Record: found
Abstract: found
Article: not found

A survey on deep learning in medical image analysis

Geert Litjens, Arnaud Arindra Adiyoso Setio, Thijs Kooi … (2017)

Deep learning algorithms, in particular convolutional networks, have rapidly become a methodology of choice for analyzing medical images. This paper reviews the major deep learning concepts pertinent to medical image analysis and summarizes over 300 contributions to the field, most of which appeared in the last year. We survey the use of deep learning for image classification, object detection, segmentation, registration, and other tasks. Concise overviews are provided of studies per application area: neuro, retinal, pulmonary, digital pathology, breast, cardiac, abdominal, musculoskeletal. We end with a summary of the current state-of-the-art, a critical discussion of open challenges and directions for future research.

0 comments Cited 2186 times – based on 0 reviews      Review now

Bookmark

Record: found
Abstract: found
Article: found

Is Open Access

A survey on Image Data Augmentation for Deep Learning

Connor Shorten, Taghi Khoshgoftaar (2019)

0 comments Cited 1571 times – based on 0 reviews      Review now

Bookmark

All references

Author and article information

Contributors

Luc Lerch: URI : https://loop.frontiersin.org/people/2717491/overviewRole: Role: Role: Role: Role: Role: Role:

Lukas S. Huber: URI : https://loop.frontiersin.org/people/2798484/overviewRole: Role: Role: Role: Role: Role: Role: Role:

Amith Kamath: Role: Role: Role: Role:

Alexander Pöllinger: Role: Role:

Aurélie Pahud de Mortanges: URI : https://loop.frontiersin.org/people/1135676/overviewRole: Role:

Verena C. Obmann: Role: Role:

Florian Dammann: Role: Role:

Walter Senn: URI : https://loop.frontiersin.org/people/2820316/overviewRole: Role:

Mauricio Reyes: Role: Role: Role: Role: Role:

Journal

Journal ID (nlm-ta): Front Radiol

Journal ID (iso-abbrev): Front Radiol

Journal ID (publisher-id): Front. Radiol.

Title: Frontiers in Radiology

Publisher: Frontiers Media S.A.

ISSN (Electronic): 2673-8740

Publication date (Electronic): 19 December 2024

Publication date Collection: 2024

Volume: 4

Electronic Location Identifier: 1420545

Affiliations

[ ¹ ]Medical Image Analysis Group, ARTORG Centre for Biomedical Research, University of Bern , Bern, Switzerland

[ ² ]Computational Neuroscience Group, Department of Physiology, University of Bern , Bern, Switzerland

[ ³ ]Cognition, Perception and Research Methods, Department of Psychology, University of Bern , Bern, Switzerland

[ ⁴ ]Neural Information Processing Group, Department of Computer Science, University of Tübingen , Tübingen, Germany

[ ⁵ ]Department of Diagnostic, Interventional, and Pediatric Radiology, Inselspital Bern, University of Bern , Bern, Switzerland

[ ⁶ ]Center for Artificial Intelligence in Medicine, University of Bern , Bern, Switzerland

[ ⁷ ]ARTORG Center for Biomedical Engineering Research, University of Bern , Bern, Switzerland

[ ⁸ ]Department of Radiation Oncology, University Hospital Bern, University of Bern , Bern, Switzerland

Author notes

Edited by: Curtise K.C. Ng, Curtin University, Australia

Reviewed by: Xi Wang, The Chinese University of Hong Kong, China

Vincent W.S. Leung, Hong Kong Polytechnic University, Hong Kong SAR, China

[* ] Correspondence: Luc Lerch luc.lerch@ 123456unibe.ch

[ † ]

These authors have contributed equally to this work and share first authorship

[ ‡ ]

Present Address: Verena C. Obmann, Department of Radiology, Cantonal Hospital of Zug, Baar, Switzerland

Article

DOI: 10.3389/fradi.2024.1420545

PMC ID: 11696537

SO-VID: 6c29387e-2653-44d9-8c5a-679d44a9ba69

License:

This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.

History

Date received : 20 April 2024

Date accepted : 22 November 2024

Page count

Figures: 4, Tables: 2, Equations: 2, References: 49, Pages: 11, Words: 0

Funding

Funded by: Swiss National Science Foundation

Award ID: P000PS_214659

The author(s) declare financial support was received for the research, authorship, and/or publication of this article. The research of Lukas S. Huber is funded by the Swiss National Science Foundation (project number: P000PS_214659). This work was supported by the National Science Foundation Switzerland, project no. 205320_212939.39.

Custom metadata

section-at-acceptance Artificial Intelligence in Radiology

Keywords: deep learning,robustness,ultrasound,breast cancer,generative adversarial network,convolutional neural network

Data availability:

Keywords: deep learning, robustness, ultrasound, breast cancer, generative adversarial network, convolutional neural network

Comments

Comment on this article

scite_

Smart Citations

Citing PublicationsSupportingMentioningContrasting

View Citations

See how this article has been cited at scite.ai

scite shows how a scientific paper has been cited by providing the context of the citation, a classification describing whether it supports, mentions, or contrasts the cited claim, and a label indicating in which section the citation was made.

Most referenced authors 797

See all reference authors

DreamOn: a data augmentation strategy to narrow the robustness gap between expert radiologists and deep learning classifiers

Read this article at

Abstract

Purpose

Materials and methods

Results

Conclusions

Related collections

Computer Vision, Deep Learning, Deep Reinforcement Learning, IoT

Most cited references 43

The Measurement of Observer Agreement for Categorical Data

A survey on deep learning in medical image analysis

A survey on Image Data Augmentation for Deep Learning

Author and article information

Contributors

Journal

Affiliations

Author notes

Article

History

Page count

Funding

Categories

Custom metadata

Comments

Comment on this article

Similar content 70

Most referenced authors 797