Gallery

Azadeh Bayani defended her Thesis titled “Designing a clinical decision support system for grading non-alcoholic fatty liver disease based on ultrasound images fuzzy processing”

NAFLD is among prevalent liver diseases in the world. Ultrasound images are commonly used; however, due to the low quality of images and the dependency of results on sonographer’s interpretation, the accuracy of diagnosis decreases. Therefore, the application of a decision support system in the diagnosis and grading of a disease has been considered in this study. After applying pre-processing algorithms, a fuzzy inference system was designed in MatLab based on the images’ texture characteristics. Different stages of the disease were scored with regard to the features of images .Different parts of images like liver brightness, diaphragm recognizable, and the clarity of blood vessels were scored as A, B, and C, respectively, and the fuzzy system output consisted of different grades of illness was tested with kappa coefficient compared with the score given to images by the radiologist. The closeness of the system output to the radiologist’s view, revealed the system’s power to diagnose and grade the disease. The technique employed in this research can be used in future studies for other diffuse liver diseases which are identified by ultrasound or other radiographic techniques.

Supervisors: L.Shahmoradi, Assistant Professor

M. Langarizadeh, Assistant Professor

Advisors: A. Radmard, Assistant Professor

Reviewers: R. Safdari, Professor

V. Changizi, Professor