Department of Medicine and Surgery, Obstetrics and Gynecology Unit, University of Parma, Parma, Italy.
National Research Council, Institute of Clinical Physiology, Lecce, Italy.
Eur J Obstet Gynecol Reprod Biol. 2024 Oct;301:147-153. doi: 10.1016/j.ejogrb.2024.08.012. Epub 2024 Aug 9.
To develop a deep learning (DL)-model using convolutional neural networks (CNN) to automatically identify the fetal head position at transperineal ultrasound in the second stage of labor.
Prospective, multicenter study including singleton, term, cephalic pregnancies in the second stage of labor. We assessed the fetal head position using transabdominal ultrasound and subsequently, obtained an image of the fetal head on the axial plane using transperineal ultrasound and labeled it according to the transabdominal ultrasound findings. The ultrasound images were randomly allocated into the three datasets containing a similar proportion of images of each subtype of fetal head position (occiput anterior, posterior, right and left transverse): the training dataset included 70 %, the validation dataset 15 %, and the testing dataset 15 % of the acquired images. The pre-trained ResNet18 model was employed as a foundational framework for feature extraction and classification. CNN was trained to differentiate between occiput anterior (OA) and non-OA positions, CNN classified fetal head malpositions into occiput posterior (OP) or occiput transverse (OT) position, and CNN classified the remaining images as right or left OT. The DL-model was constructed using three convolutional neural networks (CNN) working simultaneously for the classification of fetal head positions. The performance of the algorithm was evaluated in terms of accuracy, sensitivity, specificity, F1-score and Cohen's kappa.
Between February 2018 and May 2023, 2154 transperineal images were included from eligible participants across 16 collaborating centers. The overall performance of the model for the classification of the fetal head position in the axial plane at transperineal ultrasound was excellent, with an of 94.5 % (95 % CI 92.0--97.0), a sensitivity of 95.6 % (95 % CI 96.8-100.0), a specificity of 91.2 % (95 % CI 87.3-95.1), a F1-score of 0.92 and a Cohen's kappa of 0.90. The best performance was achieved by the CNN - OA position vs fetal head malpositions - with an accuracy of 98.3 % (95 % CI 96.9-99.7), followed by CNN - OP vs OT positions - with an accuracy of 93.9 % (95 % CI 89.6-98.2), and finally, CNN - right vs left OT position - with an accuracy of 91.3 % (95 % CI 83.5-99.1).
We have developed a DL-model capable of assessing fetal head position using transperineal ultrasound during the second stage of labor with an excellent overall accuracy. Future studies should validate our DL model using larger datasets and real-time patients before introducing it into routine clinical practice.
利用卷积神经网络（CNN）开发一种深度学习（DL）模型，以便在第二产程的经会阴超声中自动识别胎儿头部位置。
前瞻性、多中心研究，纳入第二产程的单胎、足月、头位妊娠。我们使用经腹超声评估胎儿头部位置，随后使用经会阴超声获得胎儿头部的轴面图像，并根据经腹超声结果进行标记。超声图像被随机分配到三个数据集，每个数据集包含相似比例的每种胎儿头部位置（枕前位、后位、右横位和左横位）的图像：训练数据集包括 70%，验证数据集包括 15%，测试数据集包括 15%的采集图像。使用预先训练的 ResNet18 模型作为特征提取和分类的基础框架。CNN 用于区分枕前位（OA）和非-OA 位置，CNN 将胎儿头位不正分类为枕后位（OP）或枕横位（OT），CNN 将其余图像分类为右或左 OT。该 DL 模型使用三个同时工作的卷积神经网络（CNN）构建，用于分类胎儿头部位置。该算法的性能通过准确性、敏感性、特异性、F1 评分和 Cohen's kappa 进行评估。
2018 年 2 月至 2023 年 5 月，来自 16 个合作中心的 2154 名合格参与者共纳入了 2154 次经会阴图像。该模型在经会阴超声评估胎儿头部在轴面位置的总体性能非常出色，其准确率为 94.5%（95%CI 92.0-97.0），敏感性为 95.6%（95%CI 96.8-100.0），特异性为 91.2%（95%CI 87.3-95.1），F1 评分为 0.92，Cohen's kappa 为 0.90。性能最佳的是 CNN-OA 位置与胎儿头位不正的分类，准确率为 98.3%（95%CI 96.9-99.7），其次是 CNN-OP 与 OT 位置的分类，准确率为 93.9%（95%CI 89.6-98.2），最后是 CNN-右与左 OT 位置的分类，准确率为 91.3%（95%CI 83.5-99.1）。
我们已经开发出一种能够使用第二产程经会阴超声评估胎儿头部位置的 DL 模型，其整体准确性非常出色。未来的研究应在引入常规临床实践之前，使用更大的数据集和实时患者对我们的 DL 模型进行验证。
