Publication: Indeks barah payudara menggunakan rangkaian neural hmlp
Date
2005-03-01
Authors
Syed Radzuan, Syed Firdaus
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Breast cancer is one of the major disease that contribute death among women around the globe nowadays. To solve this matter, breast cancer index is introduced. This software calculate the probability of someone being affected by breast cancer. Hybrid multilayered perceptron (HMLP) neural network is used to diagnosed and checked the data got from researchers. Besides that, by combining HMLP together with modified recursive prediction error (MRPE), this network will choose the optimum weight to get precise calculation and near to the real value. HMLP network is used because of its high confidence percentage and high accuracy if compared to other neural network. On top of that, this network is still new and the latest in neural network world. All the algorithms are programmed using Borland C++ Builder software. This software is an interface for the user who wishes to use this breast cancer software.