Brain Tumor Classification From MRI Images Using Machine Learning - Chapter 1 Introduction

Aug 3, 2026

Upload by Rowan

Brain tumor detection is critical for early diagnosis and effective treatment planning, especially because malignant brain tumors have a limited five-year relative survival rate. MRI-based pipelines require extracting discriminative characteristics across different imaging modalities, followed by machine learning classification. The project preprocesses two public datasets, trains multiple models, and evaluates accuracies. DenseNet, ResNet, EfficientNet, and VGG-16 are compared, with tumor presence and tumor-type classification, then ensembled to improve multiclass performance.

LanguageEnglish

CategoryHealthcare

FormatPDF

Pages38

Reading Time1 h 36 min

AccessFree to download

Chapter 1 Introduction

Brain Conditions

1.1 Brain Tumor

1.1.1 Types of Brain Tumor

BRAIN TUMOR CLASSIFICATION FROM MRI IMAGES USING MACHINE LEARNING Vidhyapriya Ranganathan* Celshiya Udaiyar*, Jaisree Jayanth*, Meghaa P V*, Srija B*, Uthra S* * Department of Biomedical Engineering, PSG College of Technology, India Abstract: Brain tumor is a life-threatening problem and hampers the normal functioning of the human body. The average five-year relative survival rate for malignant brain tumors is 35.6 percent. For proper diagnosis and efficient treatment planning, it is necessary to detect the brain tumor in early stages. Due to advancement in medical imaging technology, the brain images are taken in different modalities. The ability to extract relevant characteristics from magnetic resonance imaging (MRI) scans is a crucial step for brain tumor classifiers. Several studies have proposed various strategies to extract relevant features from different modalities of MRI to predict the growth of abnormal tumors. Most techniques used conventional methods of image processing for feature extraction and machine learning for classification. More recently, the use of deep learning algorithms in medical imaging has resulted in significant improvements in the classification and diagnosis of brain tumors. Since tumors are located at different regions of the brain, localizing the tumor and classifying it to a particular category is a challenging task. The objective of this project is to develop a predictive system for brain tumor detection using machine learning(ensembling) . The datasets Brain Tumor MRI Images for Brain Tumor Detection(named as ‘Dataset 1’) and Brain Tumor Classification (MRI)(named as ‘Dataset 2’) are collected and are preprocessed. The datasets are trained in various machine learning models and their accuracies are evaluated. Dataset1 is trained in models DenseNet, ResNet, EfficientNet and VGG-16 and Dataset2 is trained in DenseNet, ResNet, EfficientNet and ViT models. Models trained with Dataset1 will predict the presence or absence of a tumor whereas the models trained with Dataset 2 will classify the types of tumors. VGG-16 produced higher accuracy compared to all other models trained with Dataset1 and DenseNet produced higher accuracy with Dataset2 . These two models are ensembled to further increase the accuracy of multiclass brain tumor classification CHAPTER 1 INTRODUCTION The brain is one of the largest and most complex organs in the human body. Figure 1.1. Shows an image of the Brain. It is made up of more than 100 billion nerves that communicate in trillions of connections called synapses. The brain is made up of specialized areas that are mentioned below: ● The cortex is the outermost layer of brain cells. Thinking and voluntary movements begin in the cortex. ● The brain stem is between the spinal cord and the rest of the brain. Basic functions like breathing and sleep are controlled here. ● The basal ganglia are a cluster of structures in the center of the brain. They coordinate messages between multiple areas of the brain. ● The cerebellum is at the base and the back of the brain. It is responsible for coordination and balance. Fig. 1.1. Brain Image Brain Conditions 1. Headache: There are many types of headaches; some are serious but most are not and are generally treated with painkillers. 2. Stroke : Blood flow and oxygen are suddenly interrupted to an area of brain tissue, which then dies. A blood clot, or bleeding in the brain, are the cause of most strokes. Introduction Chapter 1 3. Brain tumor: Any abnormal growth of tissue inside the brain. Whether malignant (cancer) or benign, brain tumors usually cause problems by the pressure they exert on the normal brain. 4. Glioblastoma: An aggressive, malignant brain tumor (cancer) . Brain glioblastomas progress rapidly and are very difficult to cure. 5. Parkinson's disease: Nerves in a central area of the brain degenerate slowly, causing problems with movement and coordination. A tremor of the hands is a c

FAQs

Why is early brain tumor detection important in this project?

Early detection supports proper diagnosis and efficient treatment planning. Brain tumors can severely disrupt normal body functioning.

How are the datasets used for different prediction tasks?

Dataset 1 is trained to predict tumor presence or absence. Dataset 2 is trained to classify tumor types.

Which model results are highlighted before ensembling?

VGG-16 achieves higher accuracy among models trained on Dataset 1, while DenseNet performs best among models trained on Dataset 2. These two models are ensembled to increase multiclass accuracy.

Aug 3, 2026

Upload by Rowan