Intelligent Decision Support System for Fetal Delivery using Soft Computing Techniques

Intelligent Decision Support System for Fetal Delivery using Soft Computing Techniques

R. R. Janghel (Indian Institute of Information Technology and Management Gwalior, India), Anupam Shukla (Indian Institute of Information Technology and Management Gwalior, India) and Ritu Tiwari (Indian Institute of Information Technology and Management Gwalior, India)
Copyright: © 2013 |Pages: 17
DOI: 10.4018/978-1-4666-2455-9.ch057
OnDemand PDF Download:
$30.00
List Price: $37.50

Abstract

In the present work an attempt is made to develop an intelligent Decision support system (IDSS) using the pathological attributes to predict the fetal delivery to be done normal or by surgical procedure. The pathological tests like Blood Sugar (BR), Blood pressure (BP), Resistivity Index (RI) and systolic / Diastolic (S/P) ratio will be recorded at the time of delivery. All attributes lie within a specific range for normal patient. The database consists of the attributes for cases 2 (i.e. normal and surgical delivery). Soft computing technique namely Artificial Neural Networks (ANN) are used for simulator. The attributes from dataset are used for training & testing of ANN models. Three models of ANN are trained using Back-Propagation Algorithm (BPA), Radial Basis Function Network (RBFN), Learning Vector Quantization Network (LVQN) and one hybrid approach is Adaptive Neuro-Fuzzy Inference System (ANFIS). The designing factors have been changed to get the optimized model, which gives highest recognition score. The optimized models of BPA, RBFN, LVQN and ANFIS gave accuracies of 93.75, 99.00, 87.50 and 99.50% respectively. Hence in our present research the ANFIS is the model whom efficiency and result are best .The ANFIS is the best network for mentioned problem. This system will assist doctor to take decision at the critical time of fetal delivery.
Chapter Preview
Top

Background

The expert systems are tremendously used in health care involving predicting and diagnosing a particular disease. Medical expert systems are useful in certain situations where either the case is quite complex or there are no medical experts readily available for patients (Zaheeruddin, Garima 2006,Fahad Shahbaz Khan 2008). In past decades, various methods and systems have been proposed to efficiently expertise from domain experts. The time scales into consideration, such that the variant of disease symptoms in different time scales can be precisely expressed (Gwo-Haur Hwang 2006). The use of a fuzzy expert system to predict the need for advanced neonatal resuscitation efforts in the delivery room. This system relates the maternal medical, obstetric and neonatal characteristics to the clinical conditions of the newborn (Braz J Med Biol Res 2004). The diagnosis and management of neonatal birth injuries has been developed for the use of general practitioners and obstetricians handling large number of deliveries (Veena Kumari, C.2002). The blood pressure and heart rate used for early risk assessment, hypertension complications and as a guide to preventive intervention during pregnancy (Ayala, D.E 2002). The hierarchical network where stage networks are radial basis function networks (HRBFN) and using the nearest neighbor method as decision rule instead of the approximation method used (Ky Van Ha, 1998) . Success of neural networks in medical diagnosis depends not only on the learning mechanisms and network structure but also on the data quality (Cohen, M.E.2002. Beat-to-beat interval (BBI) and symbolic blood pressure (SP) interactions in pregnant women with chronic hypertension in comparison to those in normotensive pregnant (PRE) and non-pregnant women (Voss, A. 2003).

Complete Chapter List

Search this Book:
Reset