Classification of medical images for disease diagnosis

• Main Author: Sudha Devi Manbahal Shukal.
• Other Authors: Sirajudeen s/o Gulam Razul
• Format: Final Year Project
• Language: English
• Published: 2009
• Subjects: DRNTU::Engineering::Electrical and electronic engineering::Control and instrumentation::Medical electronics
• Online Access: http://hdl.handle.net/10356/18975
• Institution: Nanyang Technological University

This report analyses how a mathematical model of the form, y=M*exp (-t/a)*(1-exp (-t/b)) +N*exp (t*c), emulating the function of the kidneys was formulated and developed using the MATLAB environment. The kidneys perform the essential function of removing waste products from the blood and regulating the water fluid levels. For the classification of the medical images, a Statistical Pattern Recognition approach, Linear Discriminant Analysis (LDA) was employed. A database comprising of over 40 renograms, taken from more than 20 renal patients was used for this project. The algorithm was first trained using the renograms, called the training set and then further developed using the test sets. The results obtained verified that the algorithms have been successfully implemented in the program written by the author.