Deep learning enables automatic quantitative assessment of puborectalis muscle and urogenital hiatus in plane of minimal hiatal dimensions.

OBJECTIVES: To measure the length, width and area of the urogenital hiatus (UH), and the length and mean echogenicity (MEP) of the puborectalis muscle (PRM), automatically and observer-independently, in the plane of minimal hiatal dimensions on transperineal ultrasound (TPUS) images, by automatic segmentation of the UH and the PRM using deep learning. METHODS: In 1318 three- and four-dimensional (3D/4D) TPUS volume datasets from 253 nulliparae at 12 and 36 weeks' gestation, two-dimensional (2D) images in the plane of minimal hiatal dimensions with the PRM at rest, on maximum contraction and on maximum Valsalva maneuver, were obtained manually and the UH and PRM were segmented manually. In total, 713 of the images were used to train a convolutional neural network (CNN) to segment automatically the UH and PRM in the plane of minimal hiatal dimensions. In the remainder of the dataset (test set 1 (TS1); 601 images, four having been excluded), the performance of the CNN was evaluated by comparing automatic and manual segmentations. The performance of the CNN was also tested on 117 images from an independent dataset (test set 2 (TS2); two images having been excluded) from 40 nulliparae at 12 weeks' gestation, which were acquired and segmented manually by a different observer. The success of automatic segmentation was assessed visually. Based on the CNN segmentations, the following clinically relevant parameters were measured: the length, width and area of the UH, the length of the PRM and MEP. The overlap (Dice similarity index (DSI)) and surface distance (mean absolute distance (MAD) and Hausdorff distance (HDD)) between manual and CNN segmentations were measured to investigate their similarity. For the measured clinically relevant parameters, the intraclass correlation coefficients (ICCs) between manual and CNN results were determined. RESULTS: Fully automatic CNN segmentation was successful in 99.0% and 93.2% of images in TS1 and TS2, respectively. DSI, MAD and HDD showed good overlap and distance between manual and CNN segmentations in both test sets. This was reflected in the respective ICC values in TS1 and TS2 for the length (0.96 and 0.95), width (0.77 and 0.87) and area (0.96 and 0.91) of the UH, the length of the PRM (0.87 and 0.73) and MEP (0.95 and 0.97), which showed good to very good agreement. CONCLUSION: Deep learning can be used to segment automatically and reliably the PRM and UH on 2D ultrasound images of the nulliparous pelvic floor in the plane of minimal hiatal dimensions. These segmentations can be used to measure reliably UH dimensions as well as PRM length and MEP. © 2018 The Authors. Ultrasound in Obstetrics & Gynecology published by John Wiley & Sons Ltd on behalf of the International Society of Ultrasound in Obstetrics and Gynecology.

Echogenicity of puborectalis muscle, cervix and vastus lateralis muscle in pregnancy in relation to mode of delivery.

OBJECTIVES: To confirm our previous observation that levator hiatal dimensions and mean echogenicity of the puborectalis muscle (MEP) are significantly different at 12 weeks' gestation in women who delivered by Cesarean section due to failure to progress compared with those who delivered vaginally. The secondary objective was to assess the association between the echogenicity of the cervix and vastus lateralis muscle and mode of delivery. METHODS: In this prospective multicenter study, 306 nulliparous women with a singleton pregnancy underwent ultrasound assessments of the pelvic floor at rest, on maximum pelvic floor muscle contraction and on maximum Valsalva maneuver, of the cervix and of the vastus lateralis muscle at 12 weeks' gestation. Dimensions of the levator hiatus, MEP and mean echogenicity of the cervix and vastus lateralis muscle were measured and compared according to mode of delivery. RESULTS: Two hundred and forty-nine women were included in the analyses. We were unable to confirm our previous finding that MEP and levator hiatal transverse diameter and area at 12 weeks' gestation are associated significantly with mode of delivery. In addition, we could not demonstrate a significant association between echogenicity of the cervix or vastus lateralis muscle and mode of delivery. Overall, MEP was a mean of 20 points lower in women in the new database as compared with the previous study, despite the use of the same ultrasound equipment. CONCLUSION: In a second, independent multicenter dataset, we were unable to confirm our previous finding that levator hiatal dimensions and MEP on pelvic floor muscle contraction are associated significantly with mode of delivery. We also found no association between echogenicity of the cervix or vastus lateralis and mode of delivery. Copyright © 2018 ISUOG. Published by John Wiley & Sons Ltd.

Structural changes in puborectalis muscle after vaginal delivery.

OBJECTIVE: To evaluate the structural composition of the puborectalis muscle before and at several timepoints after first vaginal delivery, by the use of echogenicity and area measurements, in order to explore its recovery. METHODS: Twenty nulliparous women with a singleton pregnancy underwent 3D/4D transperineal ultrasound assessments at rest, on pelvic floor muscle contraction and on Valsalva maneuver at 12 weeks' gestation and at 1 day and 1, 2, 3, 4, 6, 12, 18 and 24 weeks after vaginal delivery. The puborectalis muscle was delineated for measurements of mean echogenicity (MEP) and area (PMA). To assess changes in MEP and PMA over time, linear mixed model analysis was used. The exact number of days after delivery at each ultrasound examination was used as a covariate. RESULTS: For all timepoints after delivery, MEP was significantly decreased compared with that at 12 weeks' gestation. MEP values increased significantly over time from 1 day to 24 weeks after delivery. Although not statistically significant, a decrease in MEP was observed between 3 and 4 weeks after delivery for all maneuvers. PMA at rest and on Valsalva maneuver remained constant after delivery. CONCLUSION: When compared with measurements taken during pregnancy, a sharp decrease in MEP was observed soon after vaginal delivery, which was most likely caused by stretch trauma to the puborectalis muscle and subsequent formation of (micro) hematoma and edema. Subsequent increases in MEP may reflect the disappearance of hematoma and edema, and also the formation of connective and scar tissue. Copyright © 2018 ISUOG. Published by John Wiley & Sons Ltd.

Recovery of puborectalis muscle after vaginal delivery: an ultrasound study.

OBJECTIVES: To assess change in levator hiatal dimensions between pregnancy and different timepoints after vaginal delivery, and map recovery of the hiatus in order to contribute to secondary prevention of symptoms of pelvic floor disorders. METHODS: Twenty nulliparous women with a singleton pregnancy underwent ultrasound assessment of the pelvic floor at rest, on maximum pelvic floor muscle contraction (PFMC) and on Valsalva maneuver at 12 weeks' gestation and at 1 day and 1, 2, 3, 4, 6, 12, 18 and 24 weeks after vaginal delivery. Dimensions of the levator hiatus were measured and contractility and distensibility were calculated. The Wilcoxon signed rank test was used to compare each postpartum value with that at 12 weeks' gestation. RESULTS: Levator hiatal area at rest, on PFMC and on Valsalva maneuver was significantly increased at 1 day and at 1 and 2 weeks after vaginal delivery compared with measurements at 12 weeks' gestation. Hiatal area at rest and on PFMC from 3 weeks postpartum onward, as well as contractility from 6 weeks onward, were comparable to values at 12 weeks' gestation, whereas, a significant difference remained on Valsalva maneuver until 24 weeks after delivery. Moreover, distensibility was still increased at 24 weeks postpartum compared with measurements at 12 weeks' gestation. CONCLUSION: The puborectalis muscle has the ability to recover anatomically from a first vaginal delivery, and recovery occurs mainly during the first 3 weeks after delivery. Stretching of the puborectalis muscle, as reflected by distensibility, persisted 24 weeks after the first vaginal delivery. The data provide a better understanding of the early 'normal' regeneration process and we hypothesize that the first 3 weeks postpartum is the best window in which to start secondary prevention. Copyright © 2017 ISUOG. Published by John Wiley & Sons Ltd.

Changes in global strain of puborectalis muscle during pregnancy and postpartum.

OBJECTIVE: One of the functional parameters that can be assessed by ultrasound is muscle strain, which represents the extent of deformation of the muscle from its original shape when forces are applied to the tissue under study. The aim of this study was to evaluate the effect of pregnancy and delivery on the puborectalis muscle, by assessing changes in global strain of the muscle during and after pregnancy. METHODS: This was a secondary analysis of a prospective observational study on the association between stress urinary incontinence and levator ani muscle avulsion after delivery. Two-hundred and eighty nulliparous pregnant women underwent four-dimensional transperineal ultrasound assessments at 12 and 36 weeks of gestation, and 6 months postpartum. Tomographic ultrasound images were constructed and the puborectalis muscle was delineated by hand using programming software. After delineation, the length of the midline of the puborectalis muscle was measured at rest and during maximum pelvic floor muscle contraction, and global strain was expressed as percentile difference. Postpartum results were analyzed separately for vaginal, operative (vacuum) vaginal and Cesarean delivery. Differences in global strain of the puborectalis muscle according to mode of delivery were compared using a paired-sample t-test. The effect of partial or complete avulsion of the puborectalis muscle on postpartum strain was evaluated by ANOVA. RESULTS: In total, 254 datasets were analyzed. Global strain of the puborectalis muscle did not change during pregnancy, but after spontaneous or operative vaginal delivery the global strain diminished significantly. No significant change was observed in strain of the puborectalis muscle after Cesarean delivery. Women who suffered complete bilateral avulsion had significantly lower strain compared with women with an intact puborectalis muscle. CONCLUSION: Spontaneous or operative vacuum vaginal birth and complete bilateral avulsion of the puborectalis muscle influence negatively the strain of the puborectalis muscle. Copyright © 2017 ISUOG. Published by John Wiley & Sons Ltd.

Ultrasonographic prediction of birth weight discordance in twin pregnancies.

OBJECTIVE: This study aimed to assess the accuracy of the ultrasonographic prediction of intertwin birth weight discordance based on crown-rump length, estimated fetal weight (EFW) and abdominal circumference. METHOD: We retrospectively studied variables of prenatal surveillance of 281 twin pregnancies at the University Medical Centre Utrecht, between 2008 and 2011. RESULTS: Fifteen per cent of twins had a birth weight discordance of ≥20%. One twin more commonly had intrauterine growth restriction in the birth weight discordance group compared with the concordantly grown group (69% vs 20%, respectively). Sonographically, EFW discrepancy and increased abdominal circumference ratio were significantly correlated with birth weight discordance. The negative predictive values were high (92% and 87%, respectively). The best prediction of weight discordance was given by EFW discordance at the last ultrasound before delivery. CONCLUSION: Intertwin birth weight discordance is best predicted by an intertwin EFW discordance at the last ultrasound. If discordance is not suspected by ultrasound, the vast majority had no birth weight discordance. In case of birth weight discordance, the presence of at least one growth-restricted twin is very likely. However, because around 20% of concordantly grown twin pairs comprise at least one intrauterine growth restriction infant, one cannot rely on weight discordance alone.