Brain Tumor Classification and Detection From MRI Images Using FCM Based Segmentation and Machine Learning Models
Loading...
Date
item.page.authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
ABSTRACT
newlineBrain cancer mortality has significantly increased across all age groups due to the abnormal
newlinegrowth of tumor tissues in the brain. This uncontrolled growth leads to brain cancer, and the
newlinerising death toll over the past few decades can be attributed to the lack of timely treatment.
newlineEarly detection is key to reducing the fatality rate, as it can greatly lower the risks associated
newlinewith the disease. However, diagnosing brain tumors manually is a complex, challenging, and
newlinetime-intensive process for medical professionals. To streamline this, automatic detection and
newlinesegmentation methods have been developed, making diagnosis easier and more efficient. These
newlinetechniques are essential for reducing death rates through accurate early-stage detection. The
newlinetask of brain tumor identification and classification has become increasingly demanding and
newlinetime-consuming for specialized radiologists and pathologists. In response, automatic detection
newlineand classification systems have become vital, significantly speeding up the diagnostic process.
newlineIn recent years, machine learning classifiers have played a pivotal role in automating the
newlineclassification of brain tumors, improving both efficiency and accuracy in medical image
newlineanalysis.
newlineThe thesis presents a brain tumor detection and classification system for MRI images using
newlinedeep learning techniques. It begins with a general introduction and a detailed literature review.
newlineThe system incorporates advanced algorithms and achieves state-of-the-art performance. The
newlineresearch introduces a novel approach for brain tumor detection and classification, utilizing the
newlineImproved Fuzzy Factor Fuzzy Local Information C-Means (IFF-FLICM) segmentation and a
newlinehybrid Modified Harmony Search and Sine Cosine Algorithm (MHS-SCA)-optimized Extreme
newlineLearning Machine (ELM). The IFF-FLICM method effectively segments MRI images to detect
newlinetumor regions, while the Mexican hat wavelet transform extracts features from these segments.
newlineThese features are classified using the MHS-SCA-ELM model, with the MHS-SCA algorithm
newline