A Drug Design Using a Multi-input Multi-output Neuro-Fuzzy System
Crina Grosan
Authors:
Crina Grosan1, Ajith Abraham2 and Ştefan Ţigan3
1Department of Computer Science,
Babes-Bolyai University, Cluj-Napoca, 3400, Romania
2 IITA Professorship Program, School of Computer Science and Engineering,
Chung-Ang University, Seoul 156-756, Korea
3 University Iuliu Haţieganu, Faculty of Medicine,
Department of Biostatistics and Medical Informatics, Cluj-Napoca, Romania
This article presents a multi-input multi-output (MIMO) neuro-fuzzy
model for a pharmaceutical research problem. Designing drugs is a
current problem in the pharmaceutical research domain. By designing a
drug we mean to choose some variables of drug formulation (inputs),
for obtaining optimal characteristics of drug (outputs). To solve such
a problem we propose a neuro-fuzzy model and the performance is
compared with artificial neural networks. This research used the
experimental data obtained from the Laboratory of Pharmaceutical
Techniques of the Faculty of Pharmacy in Cluj-Napoca, Romania. The
idea is to build a multi-input -- multi-output neuro-fuzzy model
depicting the dependence between inputs and outputs. A first order
Takagi-Sugeno type fuzzy inference system is developed and it is fine
tuned using neural network learning techniques. Bootstrap techniques
were used to generate more samples of data and the number of
experimental data is reduced due to the costs and time durations of
experimentations. We obtain in this way a better estimation of some
drug parameters. Experiment results indicate that the proposed method
is efficient.