BK Tripathy1*, RK Mohanty2 and SK Parida3
1School of Information Technology and Engineering, VIT, Vellore, Tamil Nadu, India
2School of Computer Science and Engineering, VIT, Vellore, Tamil Nadu, India
3Samanta Chandra Sekhar (Autonomous) College, Puri, Odisha, India
*Corresponding Author: BK Tripathy, School of Information Technology and Engineering, VIT, Vellore, Tamil Nadu, India.
Received: October 30, 2022; Published: November 15, 2022
Brain Tumour Detection is one of the most significant and essential task in the field of medical science. Brain tumours are of two types; malignant (cancerous) and benign (non-cancerous). Early detection followed by appropriate treatment is instrumental in increasing the probability of the patient getting cured. Manual methods are mostly time consuming, inaccurate leading to incorrect diagnosis. The main aim of this paper is to develop an automated system which can accurately detect the brain tumor with the help of a modified convolutional neural network algorithm called the CNN-XGBoost, which is an integrated and XGBoost, where CNN takes care of extracting training feature and XGBoost being used as the last level detector. The proposed method extracts the brain tumor from 2D Magnetic Resonance Images (MRI) of brain by using different algorithms like clustering, feature extraction, feature selection followed by CNN-XGBoost. Generally, this work is supposed to act as an assistive technology for doctors in order to make their work smoother and accurate. CNN is being implemented through Keras and TensorFlow. The most important task of this analysis is to distinguish between normal (without tumour) and abnormal (with tumour) images based on features, textures and image abnormalities. The proposed solution achieved high accuracy of identification at low computational cost.
Keywords: Convolutional Neural Network (CNN); XGBoost; Tumor Analysis; Feature Selection; Feature Extraction; Pattern Recognition; Magnetic Resource Image; Deep Learning
Citation: BK Tripathy., et al. “Brain Tumour Detection Using Convolutional Neural Network-XGBoost". Acta Scientific Computer Sciences 4.12 (2022): 23-33.
Copyright: © 2022 BK Tripathy., et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.