RF-ShCNN: A combination of two deep models for tumor detection in brain using MRI

Title

RF-ShCNN: A combination of two deep models for tumor detection in brain using MRI

Creator

Balasubramanian S.; Mandala J.; Rao T.V.M.; Misra A.

Description

The tumor in the brain is the reason for jagged cell enlargement in the brain. Magnetic resonance imaging (MRI) is a common scheme to identify tumor existence in the brain. With these MRIs, the medical practitioner can examine and detect the abnormal growth of tissues and corroborate if the brain is influenced by a tumor or not. Due to the appearance of artificial intelligence models, the discovery of brain tumor is performed by adapting different models which thereby help in making decisions and selecting the most suitable diagnosis for patients. The main motivation of this work is to reduce the death rate. If they are not adequately treated, the survival rate of the patient decreases. The correct diagnoses help patients receive accurate treatments and survive for a long time. This paper develops a hybrid model, namely the Residual fused Shepherd convolution neural network (RF-ShCNN) for discovering tumor in the brain considering MRI. Thus, the Adaptive wiener filtering is adapted to filter image-commencing noise. Thereafter, Conditional Random Fields-Recurrent Neural Networks (CRF-RNN) are adapted for segmentation followed by the mining of essential features. Lastly, the features employed in RF-ShCNN for making effective brain tumor detection by means of MRI. Thus, the RF-ShCNN is built by unifying the deep residual network and Shepherd convolution neural network. The hybridization is done by adding a regression layer wherein the regression is fused with Fractional calculus (FC) to make effective detection. The RF-ShCNN provided better accuracy of 94%, sensitivity of 95% and specificity of 94.9%. 2023

Source

Biomedical Signal Processing and Control, Vol-88

Date

2024-01-01

Publisher

Elsevier Ltd

Subject

Adaptive wiener filter; Brain tumor; Conditional random fields; Deep residual network; Fractional calculus; MRI; Shepherd convolution neural network

Coverage

Balasubramanian S., Computer Science and Engineering, Jain (Deemed-to-be University), Faculty of Engineering and Technology, Ramanagar District, Karnataka, 562112, India; Mandala J., Department of Computer Science and Engineering, Christ (Deemed to be) University, Bangalore, India; Rao T.V.M., Department of Computer Science and Engineering, Vignan's Institute of Information Technology, Visakhapatnam, India; Misra A., School of Computer Application, Lovely Professional University, Jalandhar Delhi G.T. Road, Phagwara, Punjab, 144411, India

Rights

Restricted Access

Relation

ISSN: 17468094

Format

Online

Language

English

Type

Article

Identifier

https://doi.org/10.1016/j.bspc.2023.105656

https://www.scopus.com/inward/record.uri?eid=2-s2.0-85175637596&doi=10.1016%2fj.bspc.2023.105656&partnerID=40&md5=08264d5aa9ab240b86f54848a3353e73