Segmentation of uterus and placenta in MR images using a fully convolutional neural network

Abstract

Segmentation of the uterine cavity and placenta in fetal magnetic resonance (MR) imaging is useful for the detection of abnormalities that affect maternal and fetal health. In this study, we used a fully convolutional neural network for 3D segmentation of the uterine cavity and placenta while a minimal operator interaction was incorporated for training and testing the network. The user interaction guided the network to localize the placenta more accurately. We trained the network with 70 training and 10 validation MRI cases and evaluated the algorithm segmentation performance using 20 cases. The average Dice similarity coefficient was 92% and 82% for the uterine cavity and placenta, respectively. The algorithm could estimate the volume of the uterine cavity and placenta with average errors of 2% and 9%, respectively. The results demonstrate that the deep learning-based segmentation and volume estimation is possible and can potentially be useful for clinical applications of human placental imaging.

Placenta segmentation in ultrasound imaging: Addressing sources of uncertainty and limited field-of-view

“…Advances in placental MR imaging including microcirculation assessment ( Slator et al, 2018 ), 3D reconstruction ( Torrents-Barrena et al, 2019b ) and automatic segmentation ( Shahedi et al, 2020 ) are helping to increase the popularity of fetal MRI as a complementary modality to US in placental evaluation ( Prayer et al, 2017 ). One major advantage of MRI in placental imaging is the larger FoV it affords ( Bulas and Egloff, 2013 ).…”

SPIE Computer-Aided Diagnosis conference anniversary review

“…[70][71][72][73][74] Other abdominal topics have included bladder segmentation, small bowel analysis including segmentation and Crohn disease detection, endoscopic image analysis for polyps and cancers, liver organ and lesion segmentation, liver elastography, kidney segmentation, renal calculi detection, pancreas segmentation, pancreatic cyst classification, and uterine and placental segmentation. [75][76][77][78][79][80][81][82][83][84][85] Prostate MRI analysis, including whole gland segmentation, cancerous and noncancerous lesion detection and classification, and multiparametric and dynamic contrast-enhanced prostate MRI analysis, was also presented as part of various topics. 11,[86][87][88][89][90][91] Occasional presentations have focused on CAD in other oncologic diseases including assessment of lymphadenopathy, cervical cancer, esophageal cancer, pancreatic tumors, and multiple myeloma.…”

Evaluation of Spatial Attentive Deep Learning for Automatic Placental Segmentation on Longitudinal MRI

“…T he placenta is an intrauterine organ necessary for the maintenance of pregnancy. 1 Abnormal placental development can adversely affect maternal health and interfere with nutrient and oxygen transport to the developing fetus. Collectively, aberrant placental development contributes toward perinatal morbidity and mortality through the development of pre-eclampsia (PE) in the mother with or without fetal growth restriction.…”

“…Collectively, aberrant placental development contributes toward perinatal morbidity and mortality through the development of pre-eclampsia (PE) in the mother with or without fetal growth restriction. 1,2 MRI has been used to detect placental volume and perfusion dysfunction. 2,3 Segmentation of the placenta by MRI is the first step required for the accurate detection of volumetric abnormalities that can affect maternal and fetal health.…”

“…In recent years, machine learning and deep learning (DL) have demonstrated superior capabilities in medical image segmentation. 1,[7][8][9][10][11][12][13][14][15][16][17] For placental segmentation in MRI, Wang et al 18 have developed an online learning-based method. 19 In addition, Alansary et al 16 have implemented a three-dimensional (3D) multiscale convolutional neural network (CNN) with 3D dense conditional random fields.…”