Development Of 3D Image Vizualization And Diagnosis System For Stroke Patients Based On Magnetic Resonance Imaging

Husainiamer, Maryam Afifah (2016) Development Of 3D Image Vizualization And Diagnosis System For Stroke Patients Based On Magnetic Resonance Imaging. Project Report. UTeM, Melaka, Malaysia. (Submitted)

Text (24 Pages) Development Of 3D Image Visualization And Diagnosis System For Stroke Patients Based On Magnetic Resonance Imaging.pdf - Submitted Version Download (968kB)

Abstract

Magnetic Resonance Imaging (MRI) uses magnetic fields and radio waves to produce high quality images of the 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 abnom1al tissue. The objective of tllis project is to estimate the detail parameters for features extraction for stroke lesion, to identify the types of stroke lesion using classification proces , to visualize the stroke lesion in 3 dimensional (3D) view and to verify the performances of the technique. In this project, the image that captured by MRI will be used as the database for classification process. The region of interest (ROI) is analyzed by taking statistical calculation to obtain the value of features extraction.The different statistical values of abnom1alities are classified using rule based classification tedulique. It will show the characterization of the ROI of different types of stroke, acute or chronic stroke. After classification is done, the performances of the system are verified to ensure the expected results are achieved. The performances of this classification system are accuracy, sensitivity and specificity.After that, 3D images of the stroke lesion are visualized. The results demonstrate that 90% accuracy has been achieved for acute stroke. This will help the physicians in analyzing the brain stroke accurately and efficiently. Finally, the Graphical User Interface (GUI) was developing to make the system more attractive and used friendly.

Actions (login required)

View Item