The Detection of Brain Cancer Using Machine Learning - A Systematic Analysis of Methodology, Comparisons, Results and Challenges

Aug 6, 2026

Upload by Patrick

Brain cancer remains a deadly condition that is often not recognized early, and accurate detection is critical because tumour location, size, and shape strongly affect diagnostic reliability. Misclassification of brain tumours can lead to severe, even fatal outcomes. This systematic analysis reviews existing literature on brain tumour identification using magnetic resonance imaging, examining multiple machine-learning approaches spanning feature extraction, augmentation, segmentation, and anatomy-focused modelling.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages16

Reading Time40 min

AccessFree to download

Introduction

Brain Tumour Classification and Grading

Primary vs. Secondary Tumours

WHO Grading System

Clinical Imaging and Diagnosis Challenges

Machine Learning in Cancer Detection and Related Work

The Detection of Brain Cancer Using Machine Learning: A Systematic Analysis of Methodology, Comparisons, Results and Challenges MD Ashif Raja 1 , Anzar Hussain Lone2* , Manpreet Kaur3 , Preet Kaur 4, Rajinder Singh Sodhi 5 , Neha6 1,4 Om Sterling Global University, Hisar, India, 5J C Bose University of Science and Technology, YMCA, Faridabad, India, 1,2,3,6 Lovely Professional University, Phagwara, PunjabIndia; E-mail: [anzarhussainlone@gmail.com](anzarhussainlone@gmail.com) Abstract: Among the deadliest illnesses that is not initially recognized in the context of cancer is brain cancer. Cancerous cells are created when cells proliferate quickly and uncontrollably. The changes in tumour location, size, and shape severely impair the ability to detect brain cancers. The wrong diagnosis of a brain tumour can have terrible and fatal consequences. So there is need of high level accuracy to find the tumor classification. The categorization, segmentation, analysis, and detection of brain cancer are the main topics of this work. In order to aid researchers, This study intends to offer an extensive exploration of the existing literature regarding brain tumor identification through the utilization of magnetic resonance imaging. Three different types of medical imaging for brain cancer were subjected to machine learning approaches (feature extraction, augmentation methods, segmentation, and anatomy of brain tumours) and identifies current issues that must be resolved for various machine learning algorithms to be applied widely in the treatment of personalised brain cancer. Ultimately, this comprehensive review compiles relevant scholarly works concerning brain tumor detection, accentuating its advantages, limitations, progressions, and challenges, thus paving the way for potential forthcoming investigations. Keywords: Brain Cancer, Feature Extraction, Machine Learning Algorithm, Magnetic Resonance Imaging, Segmentation 1. INTRODUCTION In our body brain is the one of the most important and complicated organs which work with billions of cells. The brainstem, which is predominantly shielded by the skull, is made up of the cerebrum, cerebellum, and brainstem [1, 3, 4] . Uncontrolled cell division causes an abnormal cell cluster to form in proximity to or within the brain. , which show how a brain tumour grows. There are numerous ways to categorise brain tumours , such as primary, secondary, or metastatic tumours. The size, outer later, or origin of the brain tumour are typically considered when evaluating it. Primary and secondary tumours are two different categories for tumours. As a result of the primary cancer tissue's tendency to begin in the brain and remain there, the brain. From other sections of the body, the secondary tumour may impact the brain. The classification model is ranked according to the rate of cell proliferation at which it occurs. Technical names for some brain tumours include meningioma, glioblastoma, and astrcytomas. About 70% of all brain tumours are primary tumours, with the remaining 30% being secondary tumours. Just as primary tumours, which are those that first develop in the brain, are classified according to their origin, so are other tumour types. The majority of secondary tumours, on the other hand, are malignant tumours that initially develop in any other organ of the body before spreading to the brain. Fig1: Healthy and Brain with Tumor Reference: [https://www.thehealthsite.com/diseases-conditions/brain-tumour/](https://www.thehealthsite.com/diseases-conditions/brain-tumour/) These tumours may be benign or malignant (cancerous) (benign) . When compared to benign tumours, malignant brain tumours may be more dangerous because they grow quickly, invade nearby brain and spinal tissues, and provide a greater risk of death. It may be primary or secondary type. A more detailed categorization have classified these tumors into four grades. According to the WHO's grading system, gliomas ar

FAQs

Why is early detection and accurate classification of brain tumours important?

Accurate classification and grading support treatment planning and recovery. Wrong diagnosis can cause severe and fatal consequences.

Which brain-tumour types and grades are discussed in the study context?

The text distinguishes primary and secondary tumours and describes WHO grading for gliomas from I to IV.

How does the review connect brain-tumour detection with machine learning and MRI?

It summarizes literature using magnetic resonance imaging and machine-learning methods such as feature extraction, augmentation, and segmentation, while highlighting unresolved challenges for broader application.

Aug 6, 2026

Upload by Patrick