RESUMO
Introduction Speckle tracking echocardiography (STE) is a non-Doppler modality allowing the semiautomated evaluation of the fetal cardiac function by tracking the speckles of the endocardial borders. Little evidence is available on the evaluation and comparison of different software for the functional assessment of the fetal heart by means of STE. The aim of this study is to evaluate the reproducibility and agreement of two different proprietary speckle tracking software for the prenatal semi-automated assessment of the fetal cardiac function. Methods Prospective study including non-anomalous fetuses referred for different indications at two tertiary academic units in Italy (University of Parma) and Spain (University of Barcelona). 2D clips of the four-chamber view of the fetal heart were acquired by two operators using high-end ultrasound machines with a frame rate higher than 60 Hz. The stored clips were pseudo-anonymized and shared between the collaborating Units. Functional echocardiographic analyses were independently performed using the two proprietary software (TomTec GmbH and FetalHQ®) by the same operators. Inter-software reproducibility of the endocardial global longitudinal strain (EndoGLS) and fractional area change (FAC) of the left (LV) and the right ventricles (RV) and ejection fraction (EF) of the LV were evaluated by the intraclass correlation coefficient (ICC). Results 48 fetuses were included at a median of 31+2 (21+6 - 40+3) gestational weeks. Moderate reproducibility was found for the functional parameters of the LV: EndoGLS (Pearson's correlation 0.456, p<0.01; ICC 0.446, 95%CI (0.189-0.647), p<0.01); EF (Pearson's correlation 0.435, p<0.01; ICC 0.419, 95%CI (0.156-0.627), p<0.01); FAC (Person's correlation 0.484, p<0.01; ICC 0.475, 95%CI (0.223-0.667), p<0.01). On the contrary, RV functional parameters showed poor reproducibility between the two software: EndoGLS (Pearson's correlation 0.383, p=0.01; ICC 0.377, 95%CI (0.107-0.596), p<0.01) and FAC (ICC 0.284, 95%CI (0.003-0.524), p=0.02). Conclusion Our results demonstrate a moderate reproducibility of the speckle tracking analysis of the LV using TomTec GmbH and FetalHQ®, with poor reproducibility for RV analysis.
RESUMO
OBJECTIVES: 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. MATERIAL AND METHODS: 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. CNN1 was trained to differentiate between occiput anterior (OA) and non-OA positions, CNN2 classified fetal head malpositions into occiput posterior (OP) or occiput transverse (OT) position, and CNN3 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. RESULTS: 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 CNN1 - OA position vs fetal head malpositions - with an accuracy of 98.3 % (95 % CI 96.9-99.7), followed by CNN2 - OP vs OT positions - with an accuracy of 93.9 % (95 % CI 89.6-98.2), and finally, CNN3 - right vs left OT position - with an accuracy of 91.3 % (95 % CI 83.5-99.1). CONCLUSIONS: 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.