Computerized prediction system for the route of delivery (vaginal birth versus cesarean section)

Aim: To develop a supportive computerized prediction system for the route of delivery. Materials and methods: This study consisted of 2127, 3548 and 1723 deliveries for the years 1976,1986 and 1996, respectively. We have developed a supervised artificial neural network (ANN) for predicting the delivery route. We have used a back-propagation learning algorithm for this purpose. The data used for the computerized system were obtained during the admission of the patients to the delivery room at the beginning of the first stage of labor after pelvic examination and examination/evaluation of the mother and fetus. The input variables for the ANN were maternal age, gravida, parity, gestational age at birth, necessity and type of labor induction, presentation of the baby at birth, and maternal disorders and/or risk factors. The outputs of the algorithm were vaginal delivery or cesarean section (CS). Results: The CS rates in 1976,1986 and 1996 were 9.1%418.5% and 44.5%, respectively. We have demonstrated that the system's specificity and sensitivity were 97.5% and 60.9%, respectively. The false positive rate was 25%, whereas the false negative rate was 39.1%. The positive and negative predictive values were 81.8% and 93.1%, respectively. Conclusion: Our computerized system, heretofore named as the Adana System, can be used as a supportive decision-making system for predicting the delivery route. The Adana System might be a useful tool to protect physicians against adverse medicolegal issues.