Next Article in Journal
Pultruded FRP Beams with Embedded Fibre Bragg Grating Optical Sensors for Strain Measurement and Failure Detection
Next Article in Special Issue
Impact of Lung Segmentation on the Diagnosis and Explanation of COVID-19 in Chest X-ray Images
Previous Article in Journal
Comparison of Manual Wheelchair and Pushrim-Activated Power-Assisted Wheelchair Propulsion Characteristics during Common Over-Ground Maneuvers
Previous Article in Special Issue
Co-Density Distribution Maps for Advanced Molecule Colocalization and Co-Distribution Analysis
Article

Medical Augmentation (Med-Aug) for Optimal Data Augmentation in Medical Deep Learning Networks

by 1,2, 1,2, 1,2 and 1,2,3,4,*
1
Electrical, Computer and Biomedical Engineering, Ryerson University, Toronto, ON M5B 2K3, Canada
2
Institute for Biomedical Engineering, Science and Technology (iBEST) at Ryerson University & St. Michael’s Hospital, Toronto, ON M5B 1T8, Canada
3
The Keenan Research Centre for Biomedical Science, St. Michael’s Hospital, Toronto, ON M5B 1T8, Canada
4
Department of Obstetrics and Gynecology, Faculty of Medicine, University of Toronto, Toronto, ON M5G 1E2, Canada
*
Author to whom correspondence should be addressed.
Academic Editors: Sheryl Berlin Brahnam, Loris Nanni and Rick Brattin
Sensors 2021, 21(21), 7018; https://doi.org/10.3390/s21217018
Received: 12 September 2021 / Revised: 11 October 2021 / Accepted: 18 October 2021 / Published: 22 October 2021
(This article belongs to the Special Issue Medical Image Classification)
Deep learning (DL) algorithms have become an increasingly popular choice for image classification and segmentation tasks; however, their range of applications can be limited. Their limitation stems from them requiring ample data to achieve high performance and adequate generalizability. In the case of clinical imaging data, images are not always available in large quantities. This issue can be alleviated by using data augmentation (DA) techniques. The choice of DA is important because poor selection can possibly hinder the performance of a DL algorithm. We propose a DA policy search algorithm that offers an extended set of transformations that accommodate the variations in biomedical imaging datasets. The algorithm makes use of the efficient and high-dimensional optimizer Bi-Population Covariance Matrix Adaptation Evolution Strategy (BIPOP-CMA-ES) and returns an optimal DA policy based on any input imaging dataset and a DL algorithm. Our proposed algorithm, Medical Augmentation (Med-Aug), can be implemented by other researchers in related medical DL applications to improve their model’s performance. Furthermore, we present our found optimal DA policies for a variety of medical datasets and popular segmentation networks for other researchers to use in related tasks. View Full-Text
Keywords: deep learning; data augmentation; segmentation; fetal MRI; convolutional neural networks deep learning; data augmentation; segmentation; fetal MRI; convolutional neural networks
Show Figures

Figure 1

MDPI and ACS Style

Lo, J.; Cardinell, J.; Costanzo, A.; Sussman, D. Medical Augmentation (Med-Aug) for Optimal Data Augmentation in Medical Deep Learning Networks. Sensors 2021, 21, 7018. https://doi.org/10.3390/s21217018

AMA Style

Lo J, Cardinell J, Costanzo A, Sussman D. Medical Augmentation (Med-Aug) for Optimal Data Augmentation in Medical Deep Learning Networks. Sensors. 2021; 21(21):7018. https://doi.org/10.3390/s21217018

Chicago/Turabian Style

Lo, Justin, Jillian Cardinell, Alejo Costanzo, and Dafna Sussman. 2021. "Medical Augmentation (Med-Aug) for Optimal Data Augmentation in Medical Deep Learning Networks" Sensors 21, no. 21: 7018. https://doi.org/10.3390/s21217018

Find Other Styles
Note that from the first issue of 2016, MDPI journals use article numbers instead of page numbers. See further details here.

Article Access Map by Country/Region

1
Back to TopTop