Multi-Class Brain Lesion Classification Using Deep Transfer Learning With MobileNetV3

Diagnosing brain tumors is challenging for radiologists because of the significant similarities between the tumor types. Deep learning models lack sufficient data to effectively learn the patterns of different tumors, leading adopting of transfer learning as a successful approach. However, many exis...

Full description

Saved in:
Bibliographic Details
Main Authors: Ahmed Firas Majeed, Pedram Salehpour, Leili Farzinvash, Saeid Pashazadeh
Format: Article
Language:English
Published: IEEE 2024-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/10552836/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850178360910544896
author Ahmed Firas Majeed
Pedram Salehpour
Leili Farzinvash
Saeid Pashazadeh
author_facet Ahmed Firas Majeed
Pedram Salehpour
Leili Farzinvash
Saeid Pashazadeh
author_sort Ahmed Firas Majeed
collection DOAJ
description Diagnosing brain tumors is challenging for radiologists because of the significant similarities between the tumor types. Deep learning models lack sufficient data to effectively learn the patterns of different tumors, leading adopting of transfer learning as a successful approach. However, many existing models used for this purpose are complex and involve numerous parameters and layers. In this study, we employed a lightweight MobileNetV3 model to extract features, specifically designed for mobile CPU usage, to transfer knowledge. We then design our model for brain lesion classification by incorporating lightweight DepthWise and PointWise blocks. A combination of three datasets with identical image structures is utilized, and compared its classification performance with both pre-trained and fine-tuned methods. The proposed model achieves an accuracy of 91%, outperforming other pre-trained and fine-tuned methods. Furthermore, we conduct separate accuracy assessments for each dataset, demonstrating superior performance compared to existing methods. Specifically, our model achieves an accuracy of 91% on the NINS 2022 dataset and 94% on the SBE-SMU dataset.
format Article
id doaj-art-bbfd160122434afc99631cbe2201e2ee
institution OA Journals
issn 2169-3536
language English
publishDate 2024-01-01
publisher IEEE
record_format Article
series IEEE Access
spelling doaj-art-bbfd160122434afc99631cbe2201e2ee2025-08-20T02:18:46ZengIEEEIEEE Access2169-35362024-01-011215529515530810.1109/ACCESS.2024.341300810552836Multi-Class Brain Lesion Classification Using Deep Transfer Learning With MobileNetV3Ahmed Firas Majeed0Pedram Salehpour1https://orcid.org/0000-0002-1300-7848Leili Farzinvash2Saeid Pashazadeh3https://orcid.org/0000-0002-8949-9180Department of Computer Engineering, Faculty of Electrical and Computer Engineering, University of Tabriz, Tabriz, IranDepartment of Computer Engineering, Faculty of Electrical and Computer Engineering, University of Tabriz, Tabriz, IranDepartment of Computer Engineering, Faculty of Electrical and Computer Engineering, University of Tabriz, Tabriz, IranDepartment of Computer Engineering, Faculty of Electrical and Computer Engineering, University of Tabriz, Tabriz, IranDiagnosing brain tumors is challenging for radiologists because of the significant similarities between the tumor types. Deep learning models lack sufficient data to effectively learn the patterns of different tumors, leading adopting of transfer learning as a successful approach. However, many existing models used for this purpose are complex and involve numerous parameters and layers. In this study, we employed a lightweight MobileNetV3 model to extract features, specifically designed for mobile CPU usage, to transfer knowledge. We then design our model for brain lesion classification by incorporating lightweight DepthWise and PointWise blocks. A combination of three datasets with identical image structures is utilized, and compared its classification performance with both pre-trained and fine-tuned methods. The proposed model achieves an accuracy of 91%, outperforming other pre-trained and fine-tuned methods. Furthermore, we conduct separate accuracy assessments for each dataset, demonstrating superior performance compared to existing methods. Specifically, our model achieves an accuracy of 91% on the NINS 2022 dataset and 94% on the SBE-SMU dataset.https://ieeexplore.ieee.org/document/10552836/Brain tumor detectionpre-trained modelstransfer learningMobileNetV3SmallMRI classification
spellingShingle Ahmed Firas Majeed
Pedram Salehpour
Leili Farzinvash
Saeid Pashazadeh
Multi-Class Brain Lesion Classification Using Deep Transfer Learning With MobileNetV3
IEEE Access
Brain tumor detection
pre-trained models
transfer learning
MobileNetV3Small
MRI classification
title Multi-Class Brain Lesion Classification Using Deep Transfer Learning With MobileNetV3
title_full Multi-Class Brain Lesion Classification Using Deep Transfer Learning With MobileNetV3
title_fullStr Multi-Class Brain Lesion Classification Using Deep Transfer Learning With MobileNetV3
title_full_unstemmed Multi-Class Brain Lesion Classification Using Deep Transfer Learning With MobileNetV3
title_short Multi-Class Brain Lesion Classification Using Deep Transfer Learning With MobileNetV3
title_sort multi class brain lesion classification using deep transfer learning with mobilenetv3
topic Brain tumor detection
pre-trained models
transfer learning
MobileNetV3Small
MRI classification
url https://ieeexplore.ieee.org/document/10552836/
work_keys_str_mv AT ahmedfirasmajeed multiclassbrainlesionclassificationusingdeeptransferlearningwithmobilenetv3
AT pedramsalehpour multiclassbrainlesionclassificationusingdeeptransferlearningwithmobilenetv3
AT leilifarzinvash multiclassbrainlesionclassificationusingdeeptransferlearningwithmobilenetv3
AT saeidpashazadeh multiclassbrainlesionclassificationusingdeeptransferlearningwithmobilenetv3