Human computer interface design for neonatal intensive care with data mining

Abstract

Neonatal Intensive care Unit (NICU) is capable of caring newborns and premature babies during their first four weeks of life. This could help to improve child mortality rate because 41% of all child death is reported during their neonatal period. Hence, a state of art modern Neonatal Intensive Care Unit is a prime need of any nation in order to improve child health and subsequently for socio economic development of a country like India. A Neonatal Intensive care Unit (NICU) consist multiple life saving machines which are continuously recording vital parameters of neonates and generating vast data at very high frequency. A context aware based timely monitoring of such systems is a crucial task which could able to help medical experts for efficient decision making. Considering multiple vital parameters and taking decision in a real time situation could be more improved by having efficient decision support system based on strong human computer interaction. In this present research work a human computer interface for Neonatal Intensive Care Unit has been proposed which utilizes Support Vector Machine to classify the health condition of neonates. A prognostic Index has been proposed which could be calculated on the basis of seven important vital parameters. Based on the value of Prognostic Index clinical rules may be identified to initiate alarms about the neonate health condition.