Cardiotocograph (CTG) contains uterine contraction (UC) and fetal heart rate (FHR) signals, which is an important information of clinical pregnant woman care. National Institute of Child Health and Human Development (NICHD) is one of the reference guides for clinical care and it classified pregnant woman into three categories including I, II, and III to evaluate the status of fetus. However, when it comes to manual interpretation was time-consuming and not easy to observe the slight differences. In this study, we combined rule-based method and eXtreme Gradient Boosting (XGBoost) analysis for intrapartum cardiotocograph classification. Because the category II of NICHD is defined unknown status, XGBoost analysis was used to classify the category II into IIa and IIb, and analyze their probability of fetal distress (FD). From the clinical trial of 68 pregnant women, the results of three categories (I, II and III) were consistent and no statistical difference with the clinicians’ interpretation and the average Kappa was about 0.72. The results also indicated that the probability of FD was 28.8% and 71.2% in category IIa and IIb, respectively. It shows the proposed method has potential to provide a clinical assistant tool for pregnant women care.