*Corresponding Author: suruchi.gautam@rajdhani.du.ac.in 1 DOI: https://doi.org/10.52756/ijerr.2022.v29.001 Int. J. Exp. Res. Rev., Vol. 29: 1-9 (2022) Binary and Multi-class Classification of Brain Tumors using MRI Images Suruchi Gautam 1 *, Sweety Ahlawat 2 and Prabhat Mittal 3 1 Department of Computer Science, Rajdhani College, University of Delhi, India; 2 Department of Computer Science and Engineering, Vaish College of Engineering, Maharishi Dayanand University, India; 3 Department of Business Data Processing, Satyawati College (E.), University of Delhi, India E-mail/Orcid Id: SG, suruchi.gautam@rajdhani.du.ac.in, https://orcid.org/0000-0003-0691-6530; SA, sweetyahlawat6@gmail.com; PM, profmittal@yahoo.co.in, https://orcid.org/0000-0001-5352-7955 Introduction Medical image processing makes use of various types of scans such as CT (Computer Tomography), Ultrasound, PET (Positron Emission Tomography), MRI (Magnetic Resonance Imaging), Spectroscopy, etc. Among these, MRI is most widely used for diagnosis as it is sensitive and powerful while also being noninvasive (Badža et al., 2020; Khan et al., 2020). MRI scans provide detailed information as they use effective radio waves and magnetic fields are used to create pictures of the inside organs, effectively detecting cysts, tumors, swelling or bleeding of organs. Analysis and classification of these scans lead to the identification of any irregular growth. Early detection of abnormal tissue growth is one of the main issues in medical image processing. Precise estimation of the abnormal tissue growth aids in a better prognosis and post-operative treatment. Any disease can be cured, and patients have a higher chance of surviving with early and correct detection. The fundamental unit of the human body is a cell. Tumor formation is caused by the body's cells growing irregularly or abnormally. These tumorous regions may have different shapes and sizes. Different image intensities in the scan capture these regions. Figure 1 shows an MRI scan of a normal brain and a brain with tumorous growth. A tumor can be benign or malignant. The differentiating feature among them is their structure. While benign tumors have a uniform homogenous structure, malignant or cancerous tumors form heterogenous structures. Benign tumors are non- cancerous and can be surgically removed, as they seldom grow back. Malignant tumors, however, contain cancer cells and are a cause of much concern. These cells tend to Article History: Received: 1 st Nov., 2022 Accepted: 5 th Dec., 2022 Published: 30 th Dec., 2022 Abstract: A dangerous and potentially fatal condition is a brain tumor. Early detection of this disease is critical for determining the best course of treatment. Tumor detection and classification by human inspection is a time consuming, error-prone task involving huge amounts of data. Computer-assisted machine learning and image analysis techniques have achieved significant results in image processing. In this study, we use supervised and deep learning classifiers to detect and classify tumors using the MRI images from the BRATS 2020 dataset. At the outset, the proposed system classifies images as healthy or normal brains and brain having tumorous growth. We employ four supervised machine learning classifiers SVM, Decision tree, Naïve Bayes and Linear Regression, for the binary classification. Highest accuracy (96%) was achieved with SVM and DT, with SVM giving a better Recall rate of 98%. Thereafter, categorization of the tumor as Pituitary adenoma, Meningioma, or Glioma, is performed using supervised (SVM, DT) classifiers and a 6-layer Convolution Neural Network. CNN performs better than the other classifiers, with a 93% accuracy and 92% recall rate. The suggested system is employable as a powerful decision- support tool to assist radiologists and oncologists in clinical diagnosis without requiring invasive procedures like a biopsy. Keywords: Brain tumor classification, CNN, Decision tree, Image classification, Machine learning, Support vector machine.