Automatic placental and fetal volume estimation by a convolutional neural network

Introduction: We aimed to develop an artificial intelligence (AI) deep learning algorithm to efficiently estimate placental and fetal volumes from magnetic resonance (MR) scans. Methods: Manually annotated images from an MRI sequence was used as input to the neural network DenseVNet. We included data from 193 normal pregnancies at gestational week 27 and 37. The data were split into 163 scans for training, 10 scans for validation and 20 scans for testing. The neural network segmentations were compared to the manual annotation (ground truth) using the Dice Score Coefficient (DSC). Results: The mean ground truth placental volume at gestational week 27 and 37 was 571 cm3 (Standard Deviation (SD) 293 cm3) and 853 cm3 (SD 186 cm3), respectively. Mean fetal volume was 979 cm3 (SD 117 cm3) and 2715 cm3 (SD 360 cm3). The best fitting neural network model was attained at 22,000 training iterations with mean DSC 0.925 (SD 0.041). The neural network estimated mean placental volumes at gestational week 27-870 cm3 (SD 202 cm3) (DSC 0.887 (SD 0.034), and to 950 cm3 (SD 316 cm3) at gestational week 37 (DSC 0.896 (SD 0.030)). Mean fetal volumes were 1292 cm3 (SD 191 cm3) and 2712 cm3 (SD 540 cm3), with mean DSC of 0.952 (SD 0.008) and 0.970 (SD 0.040). The time spent for volume estimation was reduced from 60 to 90 min by manual annotation, to less than 10 s by the neural network. Conclusion: The correctness of neural network volume estimation is comparable to human performance; the efficiency is substantially improved.