Deep Fractional Max Pooling Neural Network for COVID-19 Recognition
Aim: Coronavirus disease 2019 (COVID-19) is a form of disease triggered by a new strain of coronavirus. This paper proposes a novel model termed “deep fractional max pooling neural network (DFMPNN)” to diagnose COVID-19 more efficiently.Methods: This 12-layer DFMPNN replaces max pooling (MP) and ave...
Main Authors: | , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Frontiers Media S.A.
2021-08-01
|
Series: | Frontiers in Public Health |
Subjects: | |
Online Access: | https://www.frontiersin.org/articles/10.3389/fpubh.2021.726144/full |
id |
doaj-86b93e18cc05412d84c5455b41399a83 |
---|---|
record_format |
Article |
spelling |
doaj-86b93e18cc05412d84c5455b41399a832021-08-10T07:33:13ZengFrontiers Media S.A.Frontiers in Public Health2296-25652021-08-01910.3389/fpubh.2021.726144726144Deep Fractional Max Pooling Neural Network for COVID-19 RecognitionShui-Hua Wang0Suresh Chandra Satapathy1Donovan Anderson2Shi-Xin Chen3Yu-Dong Zhang4School of Mathematics and Actuarial Science, University of Leicester, Leicester, United KingdomSchool of Computer Engineering, KIIT Deemed to University, Bhubaneswar, IndiaSchool of Mathematics and Actuarial Science, University of Leicester, Leicester, United KingdomNursing Department, The Fourth People's Hospital of Huai'an, Huai'an, ChinaSchool of Informatics, University of Leicester, Leicester, United KingdomAim: Coronavirus disease 2019 (COVID-19) is a form of disease triggered by a new strain of coronavirus. This paper proposes a novel model termed “deep fractional max pooling neural network (DFMPNN)” to diagnose COVID-19 more efficiently.Methods: This 12-layer DFMPNN replaces max pooling (MP) and average pooling (AP) in ordinary neural networks with the help of a novel pooling method called “fractional max-pooling” (FMP). In addition, multiple-way data augmentation (DA) is employed to reduce overfitting. Model averaging (MA) is used to reduce randomness.Results: We ran our algorithm on a four-category dataset that contained COVID-19, community-acquired pneumonia, secondary pulmonary tuberculosis (SPT), and healthy control (HC). The 10 runs on the test set show that the micro-averaged F1 (MAF) score of our DFMPNN is 95.88%.Discussions: This proposed DFMPNN is superior to 10 state-of-the-art models. Besides, FMP outperforms traditional MP, AP, and L2-norm pooling (L2P).https://www.frontiersin.org/articles/10.3389/fpubh.2021.726144/fullconvolutional neural networkfractional max poolingdata augmentationCOVID-19average poolingmodel averaging |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Shui-Hua Wang Suresh Chandra Satapathy Donovan Anderson Shi-Xin Chen Yu-Dong Zhang |
spellingShingle |
Shui-Hua Wang Suresh Chandra Satapathy Donovan Anderson Shi-Xin Chen Yu-Dong Zhang Deep Fractional Max Pooling Neural Network for COVID-19 Recognition Frontiers in Public Health convolutional neural network fractional max pooling data augmentation COVID-19 average pooling model averaging |
author_facet |
Shui-Hua Wang Suresh Chandra Satapathy Donovan Anderson Shi-Xin Chen Yu-Dong Zhang |
author_sort |
Shui-Hua Wang |
title |
Deep Fractional Max Pooling Neural Network for COVID-19 Recognition |
title_short |
Deep Fractional Max Pooling Neural Network for COVID-19 Recognition |
title_full |
Deep Fractional Max Pooling Neural Network for COVID-19 Recognition |
title_fullStr |
Deep Fractional Max Pooling Neural Network for COVID-19 Recognition |
title_full_unstemmed |
Deep Fractional Max Pooling Neural Network for COVID-19 Recognition |
title_sort |
deep fractional max pooling neural network for covid-19 recognition |
publisher |
Frontiers Media S.A. |
series |
Frontiers in Public Health |
issn |
2296-2565 |
publishDate |
2021-08-01 |
description |
Aim: Coronavirus disease 2019 (COVID-19) is a form of disease triggered by a new strain of coronavirus. This paper proposes a novel model termed “deep fractional max pooling neural network (DFMPNN)” to diagnose COVID-19 more efficiently.Methods: This 12-layer DFMPNN replaces max pooling (MP) and average pooling (AP) in ordinary neural networks with the help of a novel pooling method called “fractional max-pooling” (FMP). In addition, multiple-way data augmentation (DA) is employed to reduce overfitting. Model averaging (MA) is used to reduce randomness.Results: We ran our algorithm on a four-category dataset that contained COVID-19, community-acquired pneumonia, secondary pulmonary tuberculosis (SPT), and healthy control (HC). The 10 runs on the test set show that the micro-averaged F1 (MAF) score of our DFMPNN is 95.88%.Discussions: This proposed DFMPNN is superior to 10 state-of-the-art models. Besides, FMP outperforms traditional MP, AP, and L2-norm pooling (L2P). |
topic |
convolutional neural network fractional max pooling data augmentation COVID-19 average pooling model averaging |
url |
https://www.frontiersin.org/articles/10.3389/fpubh.2021.726144/full |
work_keys_str_mv |
AT shuihuawang deepfractionalmaxpoolingneuralnetworkforcovid19recognition AT sureshchandrasatapathy deepfractionalmaxpoolingneuralnetworkforcovid19recognition AT donovananderson deepfractionalmaxpoolingneuralnetworkforcovid19recognition AT shixinchen deepfractionalmaxpoolingneuralnetworkforcovid19recognition AT yudongzhang deepfractionalmaxpoolingneuralnetworkforcovid19recognition |
_version_ |
1721212477795991552 |