Title | Estimation of a Neonatal Cardiovascular Risk Score by Biomedical Discriminant Analysis |
Author(s) | R.C. Hermida, F. Aguado, J.R. Fernandez, D. Ayala, J.R. Cervilla, J.M. Fraga |
Source | Proceedings of the Fifth Annual IEEE Symposium on Computer-Based Medical Systems, Pages 467-476 |
ISBN | 0-8186-2742-5 |
Publisher | IEEE Computer Society Press, Los Alamitos, CA |
Publication Date | 14-Jun-92 |
Abstract | The contribution to risk by genetics is a primary factor for the predisposition of a newborn to elevated blood pressure (BP) later in life. An index for this factor is needed to assess, on the neonate, the success or failure of preventive interventions instituted on the pregnant women. This index could be based on characteristics of BP and heart rate (HR) variability during the first days after birth. In the search for such an index, the systolic and diastolic BP and HR of 150 newborns were automatically monitorred at about 30-minute intervals for 48 hours with a Nippon Colin (Komaki, Japan) device, starting early after birth. On the basis of questionnaires given to the parents, the newborns were assigned to a group of either a negative or positive family history of high BP, according to the absence or presence of high BP and/or related cardiovascular disease in two generations (those of the newborn's parents and grandparents). Circadian characteristics and descriptive statistics for the three circulatory variables were used for classification by a so-called "monotest," an all-subsets variable selection technique for biomedical discriminant analysis. For a particular combination of variables, the "monotest" performs as many steps of separate analyses as the total number of subjects each subject'a data being compared as a set with those of all others ("leave-one-out" approach). When the 90% range of systolic BP was used as classifier, the monotest yielded a 69% total classfication equivalent to prior criteria, the latter being based on a negative or a positive family history of high BP. The monotest complements rhythmometric procedures and defines a set of individualized criteria for risk assessment. The combined use of automatic hardware for time specified sampling with proper software for signal processing and discriminant analysis allows to recognize parameters of BP and HR variability as a source of information to be used for the establishment of early preventive measures. |