BRAIN TUMOR DETECTION USING DEEP LEARNING WITH XCEPTION-HOG MODEL AND CLASSIFICATION USING MSVM TECHNIQUE
- Version
- Download 7
- File Size 604.07 KB
- File Count 1
- Create Date 20 May 2025
- Last Updated 20 May 2025
BRAIN TUMOR DETECTION USING DEEP LEARNING WITH XCEPTION-HOG MODEL AND CLASSIFICATION USING MSVM TECHNIQUE
Authors:
Mr.Hydubaigari Mahaboob Peer 1, Dr. M.N. Giri Prasad 2, Dr. D. Vishnu Vardhan 3
1 M.Tech student , Department of ECE, JNTUACEA, Anantapur, Andhra Pradesh, India
2Adjunct Professor, Department of ECE, JNTUACEA, Anantapur, Andhra Pradesh, India
3Professor, Department of ECE, JNTUACEA, Anantapur, Andhra Pradesh, India
Abstract – Early detection and precise classification of brain tumors are critical for optimizing treatment strategies and improving patient outcomes. This research introduces an advanced methodology that integrates deep learning and machine learning techniques for brain tumor detection and classification using MRI images. The proposed approach harnesses the X-ception deep learning architecture to extract high-level, discriminative features from MRI scans, leveraging its proven efficiency in complex image classification tasks. To enrich feature representation, Histogram of Oriented Gradients (HOG) is employed, capturing detailed gradient information that characterizes tumor morphology more effectively. For the classification phase, Multiple Support Vector Machines (MSVM) are utilized, extending the traditional SVM framework to robustly handle the multi-class nature of brain tumor categorization, specifically targeting meningiomas, gliomas, and pituitary tumors. By combining deep features from X-ception and handcrafted features from HOG, the model addresses challenges such as tumor heterogeneity and imaging variability. Extensive experimental evaluations demonstrate that this hybrid feature integration significantly enhances classification accuracy over conventional methods. The findings affirm the potential of the proposed system to serve as a foundation for developing automated diagnostic tools, offering valuable support to clinicians in early diagnosis and treatment planning. This study contributes meaningfully to the evolving landscape of medical imaging and underscores the transformative role of machine learning in advancing healthcare diagnostics.
Key Words: Brain Tumor Classification, MRI, X-ception, Histogram of Oriented Gradients (HOG), Multiple Support Vector Machines (MSVM), Deep Learning, Feature Extraction
Download