Please use this identifier to cite or link to this item:
http://hdl.handle.net/10603/452572
Title: | Empirical Evaluation of Deep Learning Models for MRI based Brain Tumor Classification |
Researcher: | Tandel, Gopal Singh |
Guide(s): | Kakde,O G |
Keywords: | Computer Science Computer Science Artificial Intelligence Engineering and Technology |
University: | Visvesvaraya National Institute of Technology |
Completed Date: | 2022 |
Abstract: | newline IV newlineAbstract newlineCancer is the second leading cause of mortality in men and women across the world. newlineThe most serious concern about brain tumors is that most of the other cancers may newlinedevelop brain tumors (secondary tumors) in the metastatic stage and it is responsible newlinefor 40% of all cancer deaths. The higher mortality rate is another concern of brain newlinetumor patients (over 80%) from the incidence reported at a higher stage of the disease. newlineEarly tumor diagnosis and accurate grade estimation are highly essential for prognosis newlineand treatment planning. A lack of medical facilities and highly expensive tumor newlinediagnosis methods are the main reason for late diagnosis in low or middle economic newlinecountries like India. Many lives can be saved if the tumors can be diagnosed at an newlineearly stage of progression. The Biopsy is the gold standard method of tumor grading. newlineHowever, it has several limitations such as being an invasive procedure, it can be life threatening for the patient. Secondly, the analysis of tumor tissue samples under newlinemicroscopic examination may vary from person to person. Third, under heavy newlineworkload, this process may take a few days to a week to reach a final decision. newlineHowever, Brain tumor surgery is very complicated when tumors exist deep inside the newlinebrain. It may lead to many threats to the patients such as body paralysis, or loss of life newlinesometimes. Therefore, developing a non-invasive, less expensive, fast, and accurate newlinetumor diagnosis and grading method is the major objective of this study. MRI is a newlineradiation-free imaging technique that produces a high-quality contrast image of body newlineinsides. It is the most suitable imaging test for cancer patients to detect abnormalities. newlineThus, our study aims to propose an efficient MRI-based computer-aided diagnosis and newlinegrading system for brain tumor patients. This tool can be used as an alternative or as a newlinesecond option of biopsy for the brain tumor grading system to detect tumor severity at newlinean early stage of the disease. This method is not only |
Pagination: | 212 |
URI: | http://hdl.handle.net/10603/452572 |
Appears in Departments: | Computer Science |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
80_recommendation.pdf | Attached File | 119.54 kB | Adobe PDF | View/Open |
abstract.pdf | 9.92 kB | Adobe PDF | View/Open | |
annexture.pdf | 1.47 MB | Adobe PDF | View/Open | |
chapter1.pdf | 111.26 kB | Adobe PDF | View/Open | |
chapter2.pdf | 969.28 kB | Adobe PDF | View/Open | |
chapter3.pdf | 935.89 kB | Adobe PDF | View/Open | |
chapter4.pdf | 1.1 MB | Adobe PDF | View/Open | |
chapter5.pdf | 1.2 MB | Adobe PDF | View/Open | |
chapter6.pdf | 701.71 kB | Adobe PDF | View/Open | |
content.pdf | 90.36 kB | Adobe PDF | View/Open | |
priliminary.pdf | 129.42 kB | Adobe PDF | View/Open | |
title.pdf | 39.7 kB | Adobe PDF | View/Open |
Items in Shodhganga are licensed under Creative Commons Licence Attribution-NonCommercial-ShareAlike 4.0 International (CC BY-NC-SA 4.0).
Altmetric Badge: