An Automated Detection and Segmentation of Tumor in Brain MRI using Machine Learning Technique
Journal: International Journal of Science and Research (IJSR) (Vol.11, No. 1)Publication Date: 2022-01-05
Authors : Priyanka Bharti;
Page : 51-62
Keywords : Brain tumor; Gray Level Co-Occurrence Matrix; Hidden Markov Model; K-Mean; Magnetic Resonance Image; Principal Component Analysis; Support Vector Machine;
Abstract
Magnetic Resonance Imaging can be helpful in detecting brain tumor. It has advantages over procedures like Computed Tomography scan as it does not involve use of ionizing radiation due to which exposure of a person to such risks and related side effects can be prevented. Whatever be the technique of the imaging process, the one with the maximum accuracy must be preferred. Detection of brain tumor occurs in various stages like pre-processing, segmentation, feature extraction and classifier. For this to happen, k-mean segmentation approach is applied. Gray Level Co-Occurrence Matrix and Discrete Wavelet Transform are helpful in extraction of the tumor feature. Two types of classifiers are used for classification namely Support Vector Machine and Hidden Markov Model. Then the comparison is done based on performance parameters like sensitivity, specificity and accuracy. After calculating the results, the values of performance parameters are compared. The proposed technique has been found to perform well in terms of accuracy as compared to previous technique.
Other Latest Articles
- Labyrinthine Fistula, Complication of Chronic Otitis Media: A Case Report
- Analyzing the Factors that Influence Capital Investment Decisions: A Case Study of SME's Listed on Bombay Stock Exchange
- Peptide Based Vaccine against Human Herpes Virus-7 Using in Silico Approach
- Methicillin Resistant S. aureus (MRSA) Bacteremia: A Case Series and the Clinical Implications of Vancomycin MIC Creep
- Water Quality Assessment through Water Quality Index (WQI) Determination of Narmada River during Winter Season, at Hoshanabad City, (M. P.)
Last modified: 2022-02-15 19:04:11