K-Net-Deep joint segmentation with Taylor driving training optimization based deep learning for brain tumor classification using MRI

ABSTRACT

Globally, a huge number of people succumb to brain tumour, which is considered to be one of the lethal types of tumours. In this research, an effective brain tumour segmentation and classification approach is implemented using Deep Learning (DL) based on Magnetic Resonance Imaging (MRI). Here, the segmentation of the tumour region from the brain image using the proposed hybrid K-Net-Deep joint segmentation (Deep K-Net), wherein the segmentation results produced by K-Net and Deep joint segmentation are combined using the Ruzicka similarity. Further, a Driving Training Taylor (DTT) algorithm is presented for training the K-Net. Classification is accomplished using the Shepard Convolutional Neural Network (ShCNN) that is optimized with the help of the proposed DTT algorithm. Further, the efficiency of the DTT_ShCNN is examined based on , accuracy, True Positive Rate (TPR), and True Negative Rate (TNR) with values of 0.936, 0.943, 0.945, and 0.949, respectively.

KEYWORDS:

• K-Net
• DeepJoint segmentation
• Shepard convolutional neural network
• driving training-based optimization
• Taylor series
• Deep Learning
• Ruzicka similarity
• Magnetic Resonance Imaging

Additional information

Notes on contributors

Vadamodula Prasad

Dr Vadamodula Prasad is currently associated with Lendi Institute of Engineering and Technology as Professor in the department of computer science and engineering and has an overall academic experience of 17 years. Dr. Prasad made his significant contributions towards the development of Natural Language Processing (NLP) applications and clinical applications using Artificial Intelligence (AI), Machine Learning (ML) and Deep Learning (DL) Algorithms. Presently, he is dealing with vernacular languages using NLP via interactive bots and medical applications. Dr Prasad, developed, voice applications and chat bots using AI, ML and DL Techniques. Besides this, the contributions are also towards Image Processing for Thyroid Disease Diagnose. The research contributions are mainly on Natural Language Processing, AI Bots, Deep Learning Algorithms, Machine Learning Algorithms, Rough Set Theory and applicable Math theory. Expertise at pre-processing, feature extraction, classifier selection and classifications, audio and video frame extraction from raw videos. Sound knowledge of Programming in Python, Java Server Pages, Database analysis etc. Good Number of publications in International journals and conferences in the field of NLP, Machine Learning and Deep Learning.

Vairamuthu S

Prof. Vairmuthu S is currently working as an Associate Professor in School of Computer Science and Engineering at Vellore Institute of Technology, Vellore. He has more than 20 years of teaching and research experience. He is currently supervising students in IoT, Machine Learning and Software Engineering domains. He is a NASSCOM certified Master Trainer on Analyst Security Operations Centre (SSC/Q0909).

Selva Rani B

Prof. Selva Rani B associated with School of Information Technology and Engineering at Vellore Institute of Technology, Vellore as an Associate Professor. She has 24 years of teaching and research experience. Her research domains include Recommnedation System Design, Machine Learning and Network Security. She is a NASSCOM certified Master Trainer on Analyst Security Operations Centre (SSC/Q0909).