Vasavi G, Vaddadi Vasudha Rani, Sreenu Ponnada, Jyothi S
{"title":"A hybrid EfficientNet-DbneAlexnet for brain tumor detection using MRI images.","authors":"Vasavi G, Vaddadi Vasudha Rani, Sreenu Ponnada, Jyothi S","doi":"10.1016/j.compbiolchem.2024.108279","DOIUrl":null,"url":null,"abstract":"<p><p>The rapid growth of abnormal cells in the brain presents a serious risk to the health of humans as it can result in death. Since these tumors have a varied range of shapes, sizes, and positions, identifying Brain Tumors (BTs) is challenging. Magnetic Resonance Images (MRI) are most utilized for identifying malignant tumors. This paper develops a new approach, named EfficientNet-Deep batch normalized eLUAlexnet (EfficientNet-DbneAlexnet) for detecting BTs. Firstly, the input MRI image is transmitted for image enhancement. Here, the image is enhanced by the Piecewise Linear Transformation (PLT). After this, skull stripping is carried out, which is performed by the Fuzzy Local Information C Means (FLICM). Following this, the tumor area in the image is segmented with the help of a Projective Adversarial Network (PAN). The segmented image is later applied to the feature extraction module, wherein features like textural and statistical features are extracted. Finally, the BT detection is accomplished using the developed EfficientNet-DbneAlexnet, which is created by assimilating EfficientNet and Deep batch normalized eLUAlexnet (DbneAlexnet). The results demonstrate that EfficientNet-DbneAlexnet obtained a sensitivity of 90.36 %, accuracy of 92.77 %, and specificity of 91.82 %.</p>","PeriodicalId":93952,"journal":{"name":"Computational biology and chemistry","volume":"115 ","pages":"108279"},"PeriodicalIF":0.0000,"publicationDate":"2024-11-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Computational biology and chemistry","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1016/j.compbiolchem.2024.108279","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
Abstract
The rapid growth of abnormal cells in the brain presents a serious risk to the health of humans as it can result in death. Since these tumors have a varied range of shapes, sizes, and positions, identifying Brain Tumors (BTs) is challenging. Magnetic Resonance Images (MRI) are most utilized for identifying malignant tumors. This paper develops a new approach, named EfficientNet-Deep batch normalized eLUAlexnet (EfficientNet-DbneAlexnet) for detecting BTs. Firstly, the input MRI image is transmitted for image enhancement. Here, the image is enhanced by the Piecewise Linear Transformation (PLT). After this, skull stripping is carried out, which is performed by the Fuzzy Local Information C Means (FLICM). Following this, the tumor area in the image is segmented with the help of a Projective Adversarial Network (PAN). The segmented image is later applied to the feature extraction module, wherein features like textural and statistical features are extracted. Finally, the BT detection is accomplished using the developed EfficientNet-DbneAlexnet, which is created by assimilating EfficientNet and Deep batch normalized eLUAlexnet (DbneAlexnet). The results demonstrate that EfficientNet-DbneAlexnet obtained a sensitivity of 90.36 %, accuracy of 92.77 %, and specificity of 91.82 %.