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

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...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Justin Lo, Jillian Cardinell, Alejo Costanzo, Dafna Sussman
Formato:	article
Lenguaje:	EN
Publicado:	MDPI AG 2021
Materias:	deep learning data augmentation segmentation fetal MRI convolutional neural networks Chemical technology TP1-1185
Acceso en línea:	https://doaj.org/article/d8994bac72004a1f803e410c437872f4
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:d8994bac72004a1f803e410c437872f4
record_format	dspace
spelling	oai:doaj.org-article:d8994bac72004a1f803e410c437872f42021-11-11T19:03:17ZMedical Augmentation (Med-Aug) for Optimal Data Augmentation in Medical Deep Learning Networks10.3390/s212170181424-8220https://doaj.org/article/d8994bac72004a1f803e410c437872f42021-10-01T00:00:00Zhttps://www.mdpi.com/1424-8220/21/21/7018https://doaj.org/toc/1424-8220Deep 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.Justin LoJillian CardinellAlejo CostanzoDafna SussmanMDPI AGarticledeep learningdata augmentationsegmentationfetal MRIconvolutional neural networksChemical technologyTP1-1185ENSensors, Vol 21, Iss 7018, p 7018 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	deep learning data augmentation segmentation fetal MRI convolutional neural networks Chemical technology TP1-1185
spellingShingle	deep learning data augmentation segmentation fetal MRI convolutional neural networks Chemical technology TP1-1185 Justin Lo Jillian Cardinell Alejo Costanzo Dafna Sussman Medical Augmentation (Med-Aug) for Optimal Data Augmentation in Medical Deep Learning Networks
description	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.
format	article
author	Justin Lo Jillian Cardinell Alejo Costanzo Dafna Sussman
author_facet	Justin Lo Jillian Cardinell Alejo Costanzo Dafna Sussman
author_sort	Justin Lo
title	Medical Augmentation (Med-Aug) for Optimal Data Augmentation in Medical Deep Learning Networks
title_short	Medical Augmentation (Med-Aug) for Optimal Data Augmentation in Medical Deep Learning Networks
title_full	Medical Augmentation (Med-Aug) for Optimal Data Augmentation in Medical Deep Learning Networks
title_fullStr	Medical Augmentation (Med-Aug) for Optimal Data Augmentation in Medical Deep Learning Networks
title_full_unstemmed	Medical Augmentation (Med-Aug) for Optimal Data Augmentation in Medical Deep Learning Networks
title_sort	medical augmentation (med-aug) for optimal data augmentation in medical deep learning networks
publisher	MDPI AG
publishDate	2021
url	https://doaj.org/article/d8994bac72004a1f803e410c437872f4
work_keys_str_mv	AT justinlo medicalaugmentationmedaugforoptimaldataaugmentationinmedicaldeeplearningnetworks AT jilliancardinell medicalaugmentationmedaugforoptimaldataaugmentationinmedicaldeeplearningnetworks AT alejocostanzo medicalaugmentationmedaugforoptimaldataaugmentationinmedicaldeeplearningnetworks AT dafnasussman medicalaugmentationmedaugforoptimaldataaugmentationinmedicaldeeplearningnetworks
_version_	1718431674101923840

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

Ejemplares similares