Gradient Cat Boost Classification for Enhancing the Vital Values of Pre-Term Neonatal Using Clinical Dataset

Abstract

Childbirth is the most wonderful moment for a mother. It is not only a wonderful moment, it is a rebirth for a woman with the reperfusion of every cell  by giving rise to a new offspring .It is not only an ending of pregnancy but a beginning of a new life of a bud. In NICU, Apnea is the most highly witnessed issue which we come across. It can be solved by identifying the clinical presentation of the preterm neonate by giving proper oxygen management and drugs accordingly. Neonatal babies are monitored continuously which is not equipped with a computer-aided system so far. So they decided to bring optimised machine learning to improve the data quality and predict the missed data values. The machine learning method will predict the accurate values and helps in short time complexity to save a life properly through NICU dataset from Data warehousing storages which handles complete hospital data. Datamart is a part of data warehousing, which carries the NICU datasets. Positive and negative apnea cases can be identified through a hyperplane. Multicollinear Gradient CatBoost Classification (MGCC) method helps to predict the level of drug dosage for apnea cases and predicts the missing data of apnea cases. MGCC will handle unknown data taken in the account to retrieve as predicate data to form the decision tree through GCC. Diagnosis of neonatal diseases should be done correctly, without any false data. This proposed approach plan to predict neonatal apnea cases as well as reduce the time and enhance the performance accuracy.