Application of Machine Learning for Brain Tumor Diagnosis Using Magnetic Resonance Images - A Comparative Analysis

Aug 5, 2026

Upload by Skyler

Brain tumors arise from abnormal cellular growth and can be life-threatening, making accurate diagnosis essential. MRI is a widely used conventional imaging method, yet interpreting scans can be time-intensive for clinicians. This study evaluates three machine learning approaches for reading MRI images: convolutional neural networks (CNN), random forest (RF), and support vector machine (SVM). Using the BraTS 2018 dataset with glioma, meningioma, pituitary, and no-tumor classes, Python-based experiments report high diagnostic accuracy, with CNN achieving the best performance.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages13

Reading Time33 min

AccessFree to download

Introduction

Brain tumors and clinical importance

Imaging modalities and why MRI matters

Machine learning vs. deep learning in medical image analysis

ML classifiers for brain tumor detection

Study objectives and tumor grade system

Application of machine learning for brain tumor diagnosis using magnetic resonance images: a comparative analysis Patel Rahul kumar-Manilal 1, D. J. Shah2 1Sankalchand Patel College of Engineering, Gujarat-India 2Indrashil University, Gujarat-India ORCID: 10000-0002-6696-0480, 20000-0003-1050-2330 Received: August 06, 2023. Accepted: November 02, 2023. Published: January 01, 2024. Abstract—A brain tumor is an abnormal growth of cells that may lead to cancer. MRI scans are the conventional method of diagnosing brain tumors. This paper investigates the potential of machine learning (ML) in interpreting MRI images for brain tumors. The study described applies and evaluates three different methods. The study applied and evaluated three different methods for identifying brain tumors: a selfdefined a support vector machine (SVM), a Random forest (RF), and a convolution neural network (CNN) . The Bra-TS 2018 dataset is used in this study on MRI brain images containing images of glioma, meningioma, pituitary, and no tumors. Python 3.11 was used for interpreting MRI images for brain tumors. The accuracy of the proposed CNN, RF, and SVM were found to be 99.29%, 99.06%, and 98.36%, respectively. The CNN approach has higher accuracy than innovative techniques. Keywords: magnetic resonance imaging, support vector machine, random forest, convolutional neural network, brain tumor, machine learning. I. INTRODUCTION The human Central Nervous System (CNS) consists of the brain, which is the primary component of the human nervous system, and the spinal cord [1] . The brain is responsible for overseeing the majority of the body's basic functions, including processes such as perception, integration, organization, selection, and control. The structure of the human brain is highly complex. Finding a suitable treatment for specific CNC issues, like infections, headaches, strokes, and brain tumors, can be quite challenging [2] . A brain tumor is an aggregation of anomalous cells located within the inflexible cranium that safeguards the brain [3], [4]. Any expansion inside this limited area has the potential to result in complications. The presence of a tumor within the skull poses a substantial risk to the brain, leading to brain damage [5],[6] . Brain tumors rank as the tenth leading cause of death in both children and adults [7] . Based on their texture, location, and form, brain tumors come in various varieties, all of which have very poor survival rates [8], [9] . There are several kinds of tumors, according to "The American Association of Neurological Surgeons (AANS)," as shown in Figure 1 [10] . Figure 1: Brain Tumor Classification According to AAN. Source: Own elaboration based on contributions from [11] . The most common methods for detecting abnormalities in the brain are Computed Tomography (CT), MRI, Magnetoencephalography (MEG), and Positron Emission Tomography (PET) [12] . Due to its ability to generate a wide variety of tissue contrast for all imaging methods, MRI is commonly considered the most widely used and effective tool for detecting brain diseases [13]. MRI is the predominant medical imaging technology used to visualize particular regions of the brain and get multimodal images [14] . Trained neuroradiologists have the ability to manually segment and interpret structural MRI scans of brain tumors due to their expertise and the amount of time required for this task [15],[16] . Thus, the identification and treatment of brain tumors would be significantly improved by automated and robust segmentation of the tumors. There have been various suggestions for automatically categorizing brain tumors in recent years. They could be separated into Deep Learning (DL) and ML techniques depending on feature selection, feature fusion, and the learning process. Feature selection and extraction are critical in ML algorithms for categorization [17], [18] . Contrarily, DL techniques can be learned by taking cues from ac

FAQs

Which machine learning methods are compared for brain tumor diagnosis?

The study compares CNN, random forest (RF), and support vector machine (SVM) for interpreting MRI brain images.

What dataset and tumor classes are used in the experiments?

Experiments use the BraTS 2018 dataset, containing MRI images labeled for glioma, meningioma, pituitary, and no-tumor cases.

What accuracy results are reported for the three approaches?

The reported accuracies are 99.29% for CNN, 99.06% for RF, and 98.36% for SVM.

Aug 5, 2026

Upload by Skyler