Application of the decision tree method to optimize the diagnosis of late preeclampsia on the background of metabolic syndrome.

UNLABELLED: The problem of late gestosis is one of the important problems in modern obstetrics. Despite significant advances in the diagnosis, prevention and treatment of preeclampsia, gestosis remains a formidable disease of the second half of pregnancy.

AIM: The aim of study was to investigate the possibility of using decision tree method to optimize the diagnosis and prediction of preeclampsia on the background of metabolic syndrome.

MATERIALS AND

METHODS

We conducted a retrospective analysis of 280 birth histories, records of observation of pregnant women for three years (2016-2018). In 134 women, the pregnancy was on the background of preeclampsia and metabolic syndrome. The obtained data were subsequently available for the necessary mathematical calculation by the algorithm C 4.5 specially developed for the method of induction of the decision tree, available in the licensed RStudio environment.

RESULTS

Late gestosis on the background of metabolic syndrome complicated the course of pregnancy with the occurrence of placental insufficiency, wrapping around the fetal neck 6/8 (p = 0.75), contributes to premature birth 93/105 (p = 0.88), polyhydramnios 6/6 (p = 1).

CONCLUSIONS

The correlations of a late gestosis on the background of metabolic syndrome with placental insufficiency development are revealed. Thus, the lack of predictive ability of the above factors necessitates further research to find modern molecular genetic and immunological predictors of preeclampsia development, aimed at early diagnosis and verification of the severity of this pathology.