Deepening into the suitability of using pre-trained models of ImageNet against a lightweight convolutional neural network in medical imaging: an experimental study
Transfer learning (TL) has been widely utilized to address the lack of training data for deep learning models. Specifically, one of the most popular uses of TL has been for the pre-trained models of the ImageNet dataset. Nevertheless, although these pre-trained models have shown an effective perform...
Main Authors: | , , , , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
PeerJ Inc.
2021-09-01
|
Series: | PeerJ Computer Science |
Subjects: | |
Online Access: | https://peerj.com/articles/cs-715.pdf |
id |
doaj-466a34db76cf48eb8d3eab4706c77b31 |
---|---|
record_format |
Article |
spelling |
doaj-466a34db76cf48eb8d3eab4706c77b312021-09-30T15:05:15ZengPeerJ Inc.PeerJ Computer Science2376-59922021-09-017e71510.7717/peerj-cs.715Deepening into the suitability of using pre-trained models of ImageNet against a lightweight convolutional neural network in medical imaging: an experimental studyLaith Alzubaidi0Ye Duan1Ayad Al-Dujaili2Ibraheem Kasim Ibraheem3Ahmed H. Alkenani4Jose Santamaría5Mohammed A. Fadhel6Omran Al-Shamma7Jinglan Zhang8School of Computer Science, Queensland University of Technology, Brisbane, Queensland, AustraliaFaculty of Electrical Engineering & Computer Science, University of Missouri - Columbia, Columbia, Missouri, United StatesElectrical Engineering Technical College, Middle Technical University, Baghdad, Baghdad, IraqDepartment of Electrical Engineering, College of Engineering, University of Baghdad, Baghdad, Baghdad, IraqSchool of Computer Science, Queensland University of Technology, Brisbane, Queensland, AustraliaDepartment of Computer Science, University of Jaén, Jaén, Jaén, SpainCollege of Computer Science and Information Technology, University of Sumer, Rafia, Thi Qar, IraqAlNidhal Campus, University of Information Technology & Communications, Baghdad, Baghdad, IraqSchool of Computer Science, Queensland University of Technology, Brisbane, Queensland, AustraliaTransfer learning (TL) has been widely utilized to address the lack of training data for deep learning models. Specifically, one of the most popular uses of TL has been for the pre-trained models of the ImageNet dataset. Nevertheless, although these pre-trained models have shown an effective performance in several domains of application, those models may not offer significant benefits in all instances when dealing with medical imaging scenarios. Such models were designed to classify a thousand classes of natural images. There are fundamental differences between these models and those dealing with medical imaging tasks regarding learned features. Most medical imaging applications range from two to ten different classes, where we suspect that it would not be necessary to employ deeper learning models. This paper investigates such a hypothesis and develops an experimental study to examine the corresponding conclusions about this issue. The lightweight convolutional neural network (CNN) model and the pre-trained models have been evaluated using three different medical imaging datasets. We have trained the lightweight CNN model and the pre-trained models with two scenarios which are with a small number of images once and a large number of images once again. Surprisingly, it has been found that the lightweight model trained from scratch achieved a more competitive performance when compared to the pre-trained model. More importantly, the lightweight CNN model can be successfully trained and tested using basic computational tools and provide high-quality results, specifically when using medical imaging datasets.https://peerj.com/articles/cs-715.pdfTransfer learningDeep learningImageNetConvolutional neural networkMedical imaging |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Laith Alzubaidi Ye Duan Ayad Al-Dujaili Ibraheem Kasim Ibraheem Ahmed H. Alkenani Jose Santamaría Mohammed A. Fadhel Omran Al-Shamma Jinglan Zhang |
spellingShingle |
Laith Alzubaidi Ye Duan Ayad Al-Dujaili Ibraheem Kasim Ibraheem Ahmed H. Alkenani Jose Santamaría Mohammed A. Fadhel Omran Al-Shamma Jinglan Zhang Deepening into the suitability of using pre-trained models of ImageNet against a lightweight convolutional neural network in medical imaging: an experimental study PeerJ Computer Science Transfer learning Deep learning ImageNet Convolutional neural network Medical imaging |
author_facet |
Laith Alzubaidi Ye Duan Ayad Al-Dujaili Ibraheem Kasim Ibraheem Ahmed H. Alkenani Jose Santamaría Mohammed A. Fadhel Omran Al-Shamma Jinglan Zhang |
author_sort |
Laith Alzubaidi |
title |
Deepening into the suitability of using pre-trained models of ImageNet against a lightweight convolutional neural network in medical imaging: an experimental study |
title_short |
Deepening into the suitability of using pre-trained models of ImageNet against a lightweight convolutional neural network in medical imaging: an experimental study |
title_full |
Deepening into the suitability of using pre-trained models of ImageNet against a lightweight convolutional neural network in medical imaging: an experimental study |
title_fullStr |
Deepening into the suitability of using pre-trained models of ImageNet against a lightweight convolutional neural network in medical imaging: an experimental study |
title_full_unstemmed |
Deepening into the suitability of using pre-trained models of ImageNet against a lightweight convolutional neural network in medical imaging: an experimental study |
title_sort |
deepening into the suitability of using pre-trained models of imagenet against a lightweight convolutional neural network in medical imaging: an experimental study |
publisher |
PeerJ Inc. |
series |
PeerJ Computer Science |
issn |
2376-5992 |
publishDate |
2021-09-01 |
description |
Transfer learning (TL) has been widely utilized to address the lack of training data for deep learning models. Specifically, one of the most popular uses of TL has been for the pre-trained models of the ImageNet dataset. Nevertheless, although these pre-trained models have shown an effective performance in several domains of application, those models may not offer significant benefits in all instances when dealing with medical imaging scenarios. Such models were designed to classify a thousand classes of natural images. There are fundamental differences between these models and those dealing with medical imaging tasks regarding learned features. Most medical imaging applications range from two to ten different classes, where we suspect that it would not be necessary to employ deeper learning models. This paper investigates such a hypothesis and develops an experimental study to examine the corresponding conclusions about this issue. The lightweight convolutional neural network (CNN) model and the pre-trained models have been evaluated using three different medical imaging datasets. We have trained the lightweight CNN model and the pre-trained models with two scenarios which are with a small number of images once and a large number of images once again. Surprisingly, it has been found that the lightweight model trained from scratch achieved a more competitive performance when compared to the pre-trained model. More importantly, the lightweight CNN model can be successfully trained and tested using basic computational tools and provide high-quality results, specifically when using medical imaging datasets. |
topic |
Transfer learning Deep learning ImageNet Convolutional neural network Medical imaging |
url |
https://peerj.com/articles/cs-715.pdf |
work_keys_str_mv |
AT laithalzubaidi deepeningintothesuitabilityofusingpretrainedmodelsofimagenetagainstalightweightconvolutionalneuralnetworkinmedicalimaginganexperimentalstudy AT yeduan deepeningintothesuitabilityofusingpretrainedmodelsofimagenetagainstalightweightconvolutionalneuralnetworkinmedicalimaginganexperimentalstudy AT ayadaldujaili deepeningintothesuitabilityofusingpretrainedmodelsofimagenetagainstalightweightconvolutionalneuralnetworkinmedicalimaginganexperimentalstudy AT ibraheemkasimibraheem deepeningintothesuitabilityofusingpretrainedmodelsofimagenetagainstalightweightconvolutionalneuralnetworkinmedicalimaginganexperimentalstudy AT ahmedhalkenani deepeningintothesuitabilityofusingpretrainedmodelsofimagenetagainstalightweightconvolutionalneuralnetworkinmedicalimaginganexperimentalstudy AT josesantamaria deepeningintothesuitabilityofusingpretrainedmodelsofimagenetagainstalightweightconvolutionalneuralnetworkinmedicalimaginganexperimentalstudy AT mohammedafadhel deepeningintothesuitabilityofusingpretrainedmodelsofimagenetagainstalightweightconvolutionalneuralnetworkinmedicalimaginganexperimentalstudy AT omranalshamma deepeningintothesuitabilityofusingpretrainedmodelsofimagenetagainstalightweightconvolutionalneuralnetworkinmedicalimaginganexperimentalstudy AT jinglanzhang deepeningintothesuitabilityofusingpretrainedmodelsofimagenetagainstalightweightconvolutionalneuralnetworkinmedicalimaginganexperimentalstudy |
_version_ |
1716862923439079424 |