Content uploaded by Oluwafemi Samson Balogun
Author content
All content in this area was uploaded by Oluwafemi Samson Balogun on Aug 28, 2015
Content may be subject to copyright.
A preview of the PDF is not available
ON THE USE OF DISCRIMINANT ANALYSIS IN CLASSIFICATION OF THE MODE OF DELIVERY OF AN EXPECTANT MOTHER
Abstract
The aim of the study was to obtain a discriminant function that can be used to classify the mode of delivery of pregnant women using some variables. Data from Health Records of 184 Pregnant women who delivered at the General Hospital, Wuse were used. The data consist of Mother's Weight, Height, Age and Baby's Weight, Baby's Gender and mode of delivery (Natural birth and Caesarian Section). This indicates that the Mother's Age and Mother's weight significantly affects the discrimination between the two groups. The Discriminant function D(X), which can be used in classifying the mode of delivery of women was obtained and used. The discriminant analysis gave a correct classification rate of 64.7% and misclassification rate of 35.3%.
... Maternal perinatal scale (A. Coopland) has shown high reliability and reliability in previous studies, but the scoring system has not yet been established [7][8][9][10]. The presented scale separates the factors that complicate pregnancy and childbirth, determines the degree of risk for the child during and after birth, which will allow neonatologists, pediatric anesthesiologists to predict the clinical situation and timely appoint a diagnostic algorithm and specify treatment and prevention measures. ...
Newborns with congenital defects usually have a burdened perinatal history, and their mothers' pregnancy proceeds against the background of severe complications and exacerbation of chronic pathology. In modern obstetrics, many indicative scales are used to determine the degree of risk of perinatal and maternal pathology, including a point assessment on the Coopland scale.
The aim. Conduct a clinical and epidemiological analysis of perinatal and maternal risk factors according to the A. Coopland scale and analyze the frequency of perinatal pathology in newborns with congenital defects in the early neonatal period.
Materials and methods. We retrospectively and selectively analyzed 88 birth histories of mothers of newborns with congenital malformations who received surgical treatment in the neonatal center and scored risk factors for perinatal and maternal pathology according to the A. Coopland scale.
Results. It was revealed that mothers of newborns with congenital defects were included in the groups of high (3 - 6 points) and very high (7 or more) predicted risk of perinatal and maternal pathology. The high-risk group included 28 (32%) mothers, and 60 women (68%) - the very high-risk group of perinatal and maternal pathology.
Conclusions. In women with more than 15 points on the A. Coopland scale 4.7 times more often children were born with intrauterine infection, gastrointestinal disorders, NEC and severe respiratory disorders in the early neonatal period compared with mothers who gave birth to newborns with the same gestational age and a group of low and high risk of perinatal and maternal pathology
National Drug and Enforcement Agency has been using oral evidence in the past rather than a statistical tool to classify drug offenders into Drug Peddlers and Non-Drug Peddlers.The aim of this research is to construct a Discriminant Function that can be used to classify persons for drug related offences into two groups namely: Drug Peddlers and Non-Drug Peddlers. The following variables were used: Type of Exhibit, Age, Weight of Exhibit and Gender.A discriminant function was obtained and used for classifying drug offenders into groups. The result shows that Type of Exhibit, Age, Weight of Exhibit and Gender contribute to the discriminant function. The misclassification rate obtained was 28.2%.
Mixtures of distributions, in particular the normal distribution, have been used extensively as models in a wide variety of important practical situations where the population of interest may be considered to consist of two or more subpopulations mixed in varying proportions. The problem of decomposing such a mixture of distributions is of considerable interest and utility. Two commonly used clustering methods based on maximum likelihood are considered in the context of the classification problem where observations of unknown origin belong to one of the two possible populations. The basic assumptions and associated properties of the two methods are contrasted and illustrated by a series of simulations under two different sampling schemes, namely the mixture sampling scheme and the separate sampling scheme. A case study is presented to demonstrate the basic differences between these two methods.
Method of Multivariate Analysis, Second Edition
- A C Rencher
Rencher, A.C. (2002), Method of Multivariate Analysis, Second Edition, John Wiley and Sons Inc.