Menzhulina Elena, Vitrou Juliette, Merrer Jade, Holmstrom Emilia, Amara Inesse Ait, Le Pennec Erwan, Stirnemann Julien, Ben M' Barek Imane
Department of gynecology and obstetrics - Hopital Beaujon Assistante Publique des Hôpitaux de Paris, 100 boulevard du Général Leclerc 92100 Clichy, France; Université Paris Cité, 6 rue de l'Ecole de Médecine 75006 Paris, France.
Unité d'Épidémiologie Clinique, INSERM CIC1426, Hôpital Robert Debré, APHP Paris, France.
Eur J Obstet Gynecol Reprod Biol. 2025 Apr;307:78-83. doi: 10.1016/j.ejogrb.2025.01.030. Epub 2025 Jan 29.
Cardiotocography (CTG), used during labor to assess fetal wellbeing, is subject to interobserver variability. Computerized CTG is a promising tool to improve fetal hypoxia detection.
To assess if adding clinical features improves the performance of a computerized CTG system to predict severe newborn acidemia (blood cord pH below 7.05).
A retrospective multicentric database was built using the data from two sources (the open-source CTU-UHB database and the data from Beaujon university hospital). Four CTG features were extracted from the fetal heart rate (FHR) signal (minimum and maximum value of the baseline, area covered by the accelerations and decelerations). Clinical features were also collected. Severe fetal acidemia was defined by arterial pH < 7.05 on umbilical cord sample. Risk factors for severe acidemia were sought by comparing cases with severe newborn acidemia to the rest of the cohort. We evaluated the accuracy of the model using both CTG and clinical features using area under the curve (AUC) in a cross-center, cross-validation approach.
The datasets contained 1264 cases including 100 cases with severe acidemia. In univariate analysis, hypertensive disorders and other clinical features showed no significant difference, except for meconium-stained amniotic fluid (p = 0.03). Multivariate analysis revealed that a high deceleration area (OR = 1.09 [1.04--1.11]) and apparition of meconium amniotic fluid increased the risk of newborn acidemia (OR = 2.10[1.24-3.49]). In a k-fold cross-validation approach, DeepCTG®1.5 reached an AUC of 0.77, compared to 0.74 when using CTG features only.
The CTG features have a good accuracy to predict severe newborn acidemia, confirming existing literature. Integrating clinical features tends to enhance the accuracy. Further research will aim at using more advanced machine learning models to combine the features more efficiently.
产时用于评估胎儿健康状况的胎心监护（CTG）存在观察者间差异。计算机化CTG是改善胎儿缺氧检测的一种有前景的工具。
评估添加临床特征是否能提高计算机化CTG系统预测严重新生儿酸血症（脐血pH低于7.05）的性能。
使用来自两个来源的数据（开源的CTU-UHB数据库和博若莱大学医院的数据）建立了一个回顾性多中心数据库。从胎儿心率（FHR）信号中提取了四个CTG特征（基线的最小值和最大值、加速和减速覆盖的面积）。还收集了临床特征。严重胎儿酸血症定义为脐带样本动脉pH<7.05。通过将严重新生儿酸血症病例与队列中的其他病例进行比较，寻找严重酸血症的危险因素。我们采用跨中心交叉验证方法，使用曲线下面积（AUC）评估同时使用CTG和临床特征的模型的准确性。
数据集包含1264例病例，其中100例为严重酸血症。在单变量分析中，除羊水粪染外（p = 0.03），高血压疾病和其他临床特征无显著差异。多变量分析显示，高减速面积（OR = 1.09 [l.04 - 1.11]）和羊水粪染的出现增加了新生儿酸血症的风险（OR = 2.10 [1.24 - 3.49]）。在k折交叉验证方法中，DeepCTG®1.5的AUC达到0.77，而仅使用CTG特征时为0.74。
CTG特征在预测严重新生儿酸血症方面具有良好的准确性，证实了现有文献。整合临床特征往往会提高准确性。进一步的研究将旨在使用更先进的机器学习模型更有效地结合这些特征。
