A Hybrid Deep Learning Framework for Automatic Detection of Brain Tumours Using Different Modalities

IF 5.3 3区 计算机科学 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE
Adyasha Sahu;Pradeep Kumar Das;Indraneel Paul;Sukadev Meher
{"title":"A Hybrid Deep Learning Framework for Automatic Detection of Brain Tumours Using Different Modalities","authors":"Adyasha Sahu;Pradeep Kumar Das;Indraneel Paul;Sukadev Meher","doi":"10.1109/TETCI.2024.3442889","DOIUrl":null,"url":null,"abstract":"Nowadays, deep convolutional neural networks (DCNNs) are the focus of substantial research for classification and detection applications in medical image processing. However, the limited availability and unequal data distribution of publicly available datasets impede the broad use of DCNNs for medical image processing. This work proposes a novel deep learning-based framework for efficient detection of brain tumors across different openly accessible datasets of different sizes and modalities of images. The introduction of a novel end-to-end Cumulative Learning Strategy (CLS) and Multi-Weighted New Loss (MWNL) function reduces the impact of unevenly distributed datasets. In the suggested framework, the DCNN model is incorporated with regularization, such as DropOut and DropBlock, to mitigate the problem of over-fitting. Furthermore, the suggested augmentation approach, Modified RandAugment, successfully deals with the issue of limited availability of data. Finally, the employment of K-nearest neighbor (KNN) improves the classification performance since it retains the benefits of both deep learning and machine learning. Moreover, the effectiveness of the proposed framework is also validated over small and imbalanced datasets. The proposed framework outperforms others with an accuracy of up to <inline-formula><tex-math>$ 99.70\\%$</tex-math></inline-formula>.","PeriodicalId":13135,"journal":{"name":"IEEE Transactions on Emerging Topics in Computational Intelligence","volume":"9 2","pages":"1216-1225"},"PeriodicalIF":5.3000,"publicationDate":"2024-08-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"IEEE Transactions on Emerging Topics in Computational Intelligence","FirstCategoryId":"94","ListUrlMain":"https://ieeexplore.ieee.org/document/10659042/","RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE","Score":null,"Total":0}
引用次数: 0

Abstract

Nowadays, deep convolutional neural networks (DCNNs) are the focus of substantial research for classification and detection applications in medical image processing. However, the limited availability and unequal data distribution of publicly available datasets impede the broad use of DCNNs for medical image processing. This work proposes a novel deep learning-based framework for efficient detection of brain tumors across different openly accessible datasets of different sizes and modalities of images. The introduction of a novel end-to-end Cumulative Learning Strategy (CLS) and Multi-Weighted New Loss (MWNL) function reduces the impact of unevenly distributed datasets. In the suggested framework, the DCNN model is incorporated with regularization, such as DropOut and DropBlock, to mitigate the problem of over-fitting. Furthermore, the suggested augmentation approach, Modified RandAugment, successfully deals with the issue of limited availability of data. Finally, the employment of K-nearest neighbor (KNN) improves the classification performance since it retains the benefits of both deep learning and machine learning. Moreover, the effectiveness of the proposed framework is also validated over small and imbalanced datasets. The proposed framework outperforms others with an accuracy of up to $ 99.70\%$.
使用不同模式自动检测脑肿瘤的混合深度学习框架
目前,深度卷积神经网络(deep convolutional neural networks, DCNNs)是医学图像处理中分类和检测应用的研究热点。然而,公开数据集的有限可用性和数据分布不均阻碍了DCNNs在医学图像处理中的广泛应用。这项工作提出了一种新的基于深度学习的框架,用于在不同大小和图像模式的不同开放可访问数据集上有效检测脑肿瘤。引入了一种新的端到端累积学习策略(CLS)和多加权新损失(MWNL)函数,减少了数据集分布不均匀的影响。在建议的框架中,DCNN模型与正则化(如DropOut和DropBlock)相结合,以减轻过拟合问题。此外,建议的增强方法Modified RandAugment成功地处理了数据可用性有限的问题。最后,k近邻(KNN)的使用提高了分类性能,因为它保留了深度学习和机器学习的优点。此外,在小型和不平衡的数据集上也验证了该框架的有效性。所提出的框架优于其他框架,准确率高达99.70%。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 求助全文
来源期刊
CiteScore
10.30
自引率
7.50%
发文量
147
期刊介绍: The IEEE Transactions on Emerging Topics in Computational Intelligence (TETCI) publishes original articles on emerging aspects of computational intelligence, including theory, applications, and surveys. TETCI is an electronics only publication. TETCI publishes six issues per year. Authors are encouraged to submit manuscripts in any emerging topic in computational intelligence, especially nature-inspired computing topics not covered by other IEEE Computational Intelligence Society journals. A few such illustrative examples are glial cell networks, computational neuroscience, Brain Computer Interface, ambient intelligence, non-fuzzy computing with words, artificial life, cultural learning, artificial endocrine networks, social reasoning, artificial hormone networks, computational intelligence for the IoT and Smart-X technologies.
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信