文摘
Rheumatic valvular heart disease (RVHD) and congenital heart defect (CHD) are one of the major causes of child deaths in Pakistan. According to a survey, prevalence of RVHD in Pakistan is among the highest in the world and 3.9?% of children suffer from CHD annually. If the diagnostic treatment is done earlier, then the diseases can be controlled before reaching the critical stage. Due to the lack of physicians in poor countries, there is a need for an automatic diagnostic toolkit for congenital and RVHD so that the early and accurate diagnosis of disease could be made possible even in the absence of expert physicians. In this paper, a diagnostic toolkit for rheumatic valvular and congenital heart defect is designed and developed using feed-forward back-propagation network based on symptomatic and clinical parameters. The system is built using MatLab toolkit for neural networks and graphical user interface.