Cardiotocography records fetal heart rates and their temporal relationship to uterine contractions. To identify high risk fetuses, obstetricians inspect cardiotocograms (CTGs) by eye. Therefore, CTG traces are often interpreted differently among obstetricians, resulting in inappropriate interventions. However, few studies have focused on quantitative and nonbiased algorithms for CTG evaluation. In this study, we propose a newly constructed deep neural network model (CTG-net) to detect compromised fetal status. CTG-net consists of three convolutional layers that extract temporal patterns and interrelationships between fetal heart rate and uterine contraction signals. We aimed to classify the abnormal group (umbilical artery pH < 7.20 or Apgar score at 1 min < 7) and the normal group from CTG data. We evaluated the performance of the CTG-net with the F1 score and compared it with conventional algorithms, namely, support vector machine and k-means clustering, and another deep neural network model, long short-term memory. CTG-net showed the area under the receiver operating characteristic curve of 0.73 ± 0.04, which was significantly higher than that of long short-term memory. CTG-net, a quantitative and automated diagnostic aid system, enables early intervention for putatively abnormal fetuses, resulting in a reduction in the number of cases of hypoxic injury.

Objective: To investigate the association between mid‐trimester residual cervical length (CL) and the risk of preterm birth in pregnancies after abdominal radical trachelectomy (RT).Design: Retrospective cohort study.Setting: University hospital.Population: A total of 33 deliveries after 22 weeks’ gestation in 30 women who underwent abdominal RT including prophylactic cervical cerclage and perinatal care between January 2002 and May 2016.Methods: The association between mid‐trimester residual CL (the distance between the cerclage and the external cervical os) and gestational age at delivery was investigated. Receiver–operating characteristics (ROC) curve analysis was performed to estimate the optimal cut‐off values of the mid‐trimester residual CL for the prediction of preterm birth.Main outcome measures: Preterm birth before 34 weeks’ gestation.Results: Mid‐trimester residual CL showed a significant correlation with gestational age at delivery (r = 0.36, P < 0.05). There was a significant difference in residual CL between women who did and those who did not give birth before 34 weeks (P < 0.05). Mid‐trimester residual CL < 13 mm was a good predictor of birth before 34 weeks, with a sensitivity of 67%, specificity of 75%, positive predictive value of 55% and negative predictive value of 86% (area under ROC curve, 0.75).Conclusions: Mid‐trimester residual CL is significantly correlated with gestational age at delivery. Residual CL assessment could be used to reassure physicians and women that there is only a small chance of preterm birth in pregnancies after abdominal RT.Tweetable Mid‐trimester residual cervical length is a good predictor of preterm birth after radical trachelectomy.

Despite the poor prognosis associated with myelomeningocele (MMC), the options for prenatal treatments are still limited. Recently, fetal cellular therapy has become a new option for treating birth defects, although the therapeutic effects and mechanisms associated with such treatments remain unclear. The use of human amniotic fluid stem cells (hAFSCs) is ideal with respect to immunoreactivity and cell propagation. The prenatal diagnosis of MMC during early stages of pregnancy could allow for the ex vivo proliferation and modulation of autologous hAFSCs for use in utero stem cell therapy. Therefore, we investigated the therapeutic effects and mechanisms of hAFSCs-based treatment for fetal MMC. hAFSCs were isolated as CD117-positive cells from the amniotic fluid of 15- to 17-week pregnant women who underwent amniocentesis for prenatal diagnosis and consented to this study. Rat dams were exposed to retinoic acid to induce fetal MMC and were subsequently injected with hAFSCs in each amniotic cavity. We measured the exposed area of the spinal cord and hepatocyte growth factor (HGF) levels at the lesion. The exposed spinal area of the hAFSC-treated group was significantly smaller than that of the control group. Immunohistochemical analysis demonstrated a reduction in neuronal damage such as neurodegeneration and astrogliosis in the hAFSC-treated group. Additionally, in lesions of the hAFSC-treated group, HGF expression was upregulated and HGF-positive hAFSCs were identified, suggesting that these cells migrated to the lesion and secreted HGF to suppress neuronal damage and induce neurogenesis. Therefore, in utero hAFSC therapy could become a novel strategy for fetal MMC. Stem Cells Translational Medicine 2019;8:1170–1179