Risk factors based classification for accurate prediction of the Preterm Birth

Abstract

With the advent of technological advances in the healthcare industry, predicting the labor-related complications becomes an important aspect in Gynecology and Obstetrics. It is a proactive way of preparing the patients mentally for facing any unforeseen situations which may arise due to these complications. The earlier the complications if any are detected; it is easier to prescribe the medication and the treatment to overcome the complications. Preterm Birth (PTB) is one such complication which needs a special attention and medication so that a possible PTB can be converted to a normal birth. Unfortunately, the clinical procedures like Ultrasound Scan and Swab test cannot reveal any major indicators of PTB and hence the number of spontaneous PTBs is increasing continuously. Between 1981 and 2008, PTB has increased from 9.4% to 12.3%. This is an increase of 36% in less than two decades. Hence there is a need to predict the PTB well in advance so that it helps the healthcare professionals to make decisions about the treatment. Subsequently, the expectant mother undergoes minimal or no complications of preterm labor. On the other hand, it also helps to avoid unnecessary hospitalization and treatment for women who are having a false labor pain. This study predicts the PTB by analyzing the historical data of patients who had either preterm or term birth. The results from this study show that PTB can be predicted with an accuracy of more than 98% using stacked generalization. The proposed approach helps the physicists in Gynecology and Obstetrics departments to accurately predict the PTB. Based on the prediction, the decision about the treatment to be rendered to the expectant mother to delay the birth is made on time. This, in turn, can reduce the mortality of babies due to preterm birth.