ABSTRACT

This chapter presents an algorithm called weights-and-structure-determination (WASD) for a three-layer feed-forward neuronet. It investigates an algorithm called WASD for the feed-forward neuronetfeed-forward neuronet aiming at solving the problems existing in traditional algorithms. The WASD efficacy is thus substantiated with a significant value in glomerular filtration rate (GFR) estimation of chronic kidney disease (CKD) for clinical applications. As glomerular filtration rate GFR is a commonly used index for early detection of CKD, an effective, efficient and accurate GFR-estimation method is significant for clinical applications. Along with the development of society and the changes of environment, CKD has become a public health problem which threatens people’s health and life seriously. With the consideration that GFR is a commonly used index for early detection of CKD, an effective, efficient and accurate GFR-estimation method is significant for clinical applications. Furthermore, the WASD neuronet has been applied to estimating GFR so as to address the problem of CKD for clinical applications in China.