A novel physics-based data augmentation approach for improved robust deep learning in medical imaging: lung nodule CAD false positive reduction in low-dose CT environments

M. W. Wahi-Anwar; N. Emaminejad; Y. Choi; H. G. Kim; W. Hsu; M. S. Brown; M. F. McNitt-Gray

doi:10.1117/12.2582126

16 February 2021 A novel physics-based data augmentation approach for improved robust deep learning in medical imaging: lung nodule CAD false positive reduction in low-dose CT environments

M. W. Wahi-Anwar, N. Emaminejad, Y. Choi, H. G. Kim, W. Hsu, M. S. Brown, M. F. McNitt-Gray

Author Affiliations +

Proceedings Volume 11595, Medical Imaging 2021: Physics of Medical Imaging; 115950G (2021) https://doi.org/10.1117/12.2582126
Event: SPIE Medical Imaging, 2021, Online Only

Abstract

A novel physics-based data augmentation (PBDA) is introduced, to provide a representative approach to introducing variance during training of a deep-learning model. Compared to traditional geometric-based data augmentation (GBDA), we hypothesize that PBDA will provide more realistic variation representative of potential imaging conditions that may be seen beyond the initial training data, and thereby train a more robust model (particularly in the scope of medical imaging). PBDA is tested in the context of false-positive reduction in nodule detection in low-dose lung CT and is shown to exhibit superior performance and robustness across a wide range of imaging conditions.

Conference Presentation

Citation Download Citation

M. W. Wahi-Anwar, N. Emaminejad, Y. Choi, H. G. Kim, W. Hsu, M. S. Brown, and M. F. McNitt-Gray "A novel physics-based data augmentation approach for improved robust deep learning in medical imaging: lung nodule CAD false positive reduction in low-dose CT environments", Proc. SPIE 11595, Medical Imaging 2021: Physics of Medical Imaging, 115950G (16 February 2021); https://doi.org/10.1117/12.2582126

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available