Pattern discrimination software for uncertainty reduction in neonatal cardiovascular risk assessment

Genetic risk is a primary contributing factor to the predisposition of a newborn child to elevated blood pressure later in life. To determine if there is a correlation between potential genetic risk as established by family history and measured physiological variables in the neonate, the systolic and diastolic blood pressures and heart rates of 150 newborn babies were automatically monitored at about 30-minute intervals for 48 hours with a Nippon Colin device, starting early after birth. Based on questionnaires given to the parents, the neonates were assigned to a group of either a negative or positive family history of high blood pressure. Circadian characteristics and descriptive statistics for the 3 circulatory variables were used for classification by a “monotest”, an all-subsets variable selection technique for biomedical discriminant analysis. For a particular combination of variables, the monotest performs as many separate analyses as the number of subjects, each subject´s data being compared with those of all others. When the circadian amplitudes of systolic blood pressure and heart rate and the circadian range of heart rate were used as a combined classifier, the monotest yielded a 70% classification equivalent to prior criteria, the latter being based on the family history of high blood pressure. The combined use of time-specified sampling hardware with signal processing and discriminant analysis software allows one to recognize parameters of blood pressure circadian variability as a source of information for neonatal classification according to cardiovascular risk