Brain Lesion Classification Based On Statistical Discrimination For Magnetic Resonance Imaging (MRI) Image

Syed Mohd Syafiq, Abu Bakar (2014) Brain Lesion Classification Based On Statistical Discrimination For Magnetic Resonance Imaging (MRI) Image. Project Report. UTeM, Melaka, Malaysia. (Submitted)

Text (24 pages) Brain Lesion Classification Based On Statistical Discrimination For Magnetic Resonance Imaging (MRI) Image 24 Pages.pdf - Submitted Version Download (189kB)

Abstract

Computer technology covers a wide range of medical area such as cancer research, heart diseases and brain diseases. Magnetic Resonance Imaging (MRI) uses magnetic fields and radio waves to produce high-quality two or three dimensional images of body. This imaging technique is non-radioactive and pain-free technique for visualizing and detecting brain lesion without any human involvement. It gives the detailed information regarding the abnormal tissue. The aim of this project is to classify the lesion based on statistical discrimination of abnormalities. In this project, the image that captured by MRI will be used as database that need to process for classification. From the image, the region of lesion was segment manually using handfree tool from MATLAB software to proceed the classification process. The segment region is analyzed by taking statistical calculation to obtain the value of feature extraction. The different statistical values of abnormalities are classified using ruled based classification technique. It will show the characterization of region of interest (ROI) of different types of lesion. After the classification processing done, the system is verified to ensure the expected results are achieved. The performance of this classification system was shown using graph by taking several database that is 30%, 50%, 70% and 90%. Finally, the Graphical User Interface (GUI) was developing to make the system more attractive and user friendly.

Actions (login required)

View Item