Automatic segmentation of puborectalis muscle on three-dimensional transperineal ultrasound

Applicability of Deep Learning to Dynamically Identify the Different Organs of the Pelvic Floor in the Midsagittal Plane

INTRODUCTION AND HYPOTHESIS

The objective was to create and validate the usefulness of a convolutional neural network (CNN) for identifying different organs of the pelvic floor in the midsagittal plane via dynamic ultrasound.

METHODS

This observational and prospective study included 110 patients. Transperineal ultrasound scans were performed by an expert sonographer of the pelvic floor. A video of each patient was made that captured the midsagittal plane of the pelvic floor at rest and the change in the pelvic structures during the Valsalva maneuver. After saving the captured videos, we manually labeled the different organs in each video. Three different architectures were tested-UNet, FPN, and LinkNet-to determine which CNN model best recognized anatomical structures. The best model was trained with the 86 cases for the number of epochs determined by the stop criterion via cross-validation. The Dice Similarity Index (DSI) was used for CNN validation.

RESULTS

Eighty-six patients were included to train the CNN and 24 to test the CNN. After applying the trained CNN to the 24 test videos, we did not observe any failed segmentation. In fact, we obtained a DSI of 0.79 (95% CI: 0.73 - 0.82) as the median of the 24 test videos. When we studied the organs independently, we observed differences in the DSI of each organ. The poorest DSIs were obtained in the bladder (0.71 [95% CI: 0.70 - 0.73]) and uterus (0.70 [95% CI: 0.68 - 0.74]), whereas the highest DSIs were obtained in the anus (0.81 [95% CI: 0.80 - 0.86]) and levator ani muscle (0.83 [95% CI: 0.82 - 0.83]).

CONCLUSIONS

Our results show that it is possible to apply deep learning using a trained CNN to identify different pelvic floor organs in the midsagittal plane via dynamic ultrasound.

Validation of a built-in software in automatically reconstructing the tomographic images of the levator ani muscle

INTRODUCTION AND HYPOTHESIS

Transperineal ultrasound (TPUS) is an effective tool for evaluating the integrity of the levator ani muscle (LAM). Several operating steps are required to obtain the standard multi-slice image of the LAM, which is experience dependent and time consuming. This study was aimed at evaluating the feasibility and reproducibility of the built-in software, Smart-pelvic™, in reconstructing standard tomographic images of LAM from 3D/4D TPUS volumes.

METHODS

This study was conducted at a tertiary teaching hospital, enrolling women who underwent TPUS. Tomographic images of the LAM were automatically reconstructed by Smart-pelvicTM and rated by two experienced observers as standard or nonstandard. The anteroposterior diameter (APD) of the levator hiatus was also measured on the mid-sagittal plane of the automatically and manually reconstructed images. The APD measurements of each approach were compared using Bland-Altman plots, and interclass correlation coefficient (ICC) was used to evaluate intra- and inter-observer reproducibility. Meanwhile, the time taken for the reconstruction process of both methods was also recorded.

RESULTS

The ultrasound volume of a total of 104 patients were included in this study. Using Smart-pelvicTM, the overall success rate of the tomographic image reconstruction was 98%. Regarding measurements of APD, the ICC between the automatic and manual reconstruction methods was 0.99 (0.98, 0.99). The average time taken for reconstruction per case was 2.65 ± 0.52 s and 22.08 ± 3.45 s, respectively.

CONCLUSIONS

Using Smart-pelvicTM to reconstruct tomographic images of LAM is feasible, and it can promote TPUS by reducing operator dependence and improving examination efficiency in a clinical setting.

3D ultrasound strain imaging of puborectal muscle with and without unilateral avulsion

INTRODUCTION AND HYPOTHESIS

The puborectal muscle (PRM), one of the female pelvic floor (PF) muscles, can get damaged during vaginal delivery, leading to disorders such as pelvic organ prolapse. Current diagnosis involves ultrasound (US) imaging of the female PF muscles, but functional information is limited. Previously, we developed a method for strain imaging of the PRM from US images in order to obtain functional information. In this article, we hypothesize that strain in the PRM would differ from intact to the avulsed end.

METHODS

We calculated strain in PRMs at maximum contraction, along their muscle fiber direction, from US images of two groups of women, which consisted of women with intact (n1 = 8) and avulsed PRMs (unilateral) (n2 = 10). Normalized strain ratios between both ends of the PRM (avulsed or intact) and the mid region were calculated. Subsequently, the difference in ratio between the avulsed and intact PRMs was determined.

RESULTS

We observe from the obtained results that the contraction/strain pattern of intact and undamaged PRMs is different from PRMs with unilateral avulsion. Normalized strain ratios between avulsed and intact PRMs were statistically significant (p = 0.04).

CONCLUSION

In this pilot study, we were able to show that US strain imaging of PRMs can show differences between intact PRMs and PRMs with unilateral avulsion.

Automated Segmentation of Levator Ani Muscle from 3D Endovaginal Ultrasound Images

Levator ani muscle (LAM) avulsion is a common complication of vaginal childbirth and is linked to several pelvic floor disorders. Diagnosing and treating these conditions require imaging of the pelvic floor and examination of the obtained images, which is a time-consuming process subjected to operator variability. In our study, we proposed using deep learning (DL) to automate the segmentation of the LAM from 3D endovaginal ultrasound images (EVUS) to improve diagnostic accuracy and efficiency. Over one thousand images extracted from the 3D EVUS data of healthy subjects and patients with pelvic floor disorders were utilized for the automated LAM segmentation. A U-Net model was implemented, with Intersection over Union (IoU) and Dice metrics being used for model performance evaluation. The model achieved a mean Dice score of 0.86, demonstrating a better performance than existing works. The mean IoU was 0.76, indicative of a high degree of overlap between the automated and manual segmentation of the LAM. Three other models including Attention UNet, FD-UNet and Dense-UNet were also applied on the same images which showed comparable results. Our study demonstrated the feasibility and accuracy of using DL segmentation with U-Net architecture to automate LAM segmentation to reduce the time and resources required for manual segmentation of 3D EVUS images. The proposed method could become an important component in AI-based diagnostic tools, particularly in low socioeconomic regions where access to healthcare resources is limited. By improving the management of pelvic floor disorders, our approach may contribute to better patient outcomes in these underserved areas.

Tissue Characterization of Puborectalis Muscle From 3-D Ultrasound

Pelvic floor (PF) muscles have the role of preventing pelvic organ descent. The puborectalis muscle (PRM), which is one of the female PF muscles, can be damaged during child delivery. This damage can potentially cause irreversible muscle trauma and even lead to an avulsion, which is disconnection of the muscle from its insertion point, the pubic bone. Ultrasound imaging allows diagnosis of such trauma based on comparison of geometric features of a damaged muscle with the geometric features of a healthy muscle. Although avulsion, which is considered severe damage, can be diagnosed, microdamage within the muscle itself leading to structural changes cannot be diagnosed by visual inspection through imaging only. Therefore, we developed a quantitative ultrasound tissue characterization method to obtain information on the state of the tissue of the PRM and the presence of microdamage in avulsed PRMs. The muscle was segmented as the region of interest (ROI) and further subdivided into six regions of interest (sub-ROIs). Mean echogenicity, entropy and shape parameter of the statistical distribution of gray values were analyzed on two of these sub-ROIs nearest to the bone. The regions nearest to the bones are also the most likely regions to exhibit damage in case of disconnection or avulsion. This analysis was performed for both the muscle at rest and the muscle in contraction. We found that, for PRMs with unilateral avulsion compared with undamaged PRMs, the mean echogenicity (p = 0.02) and shape parameter (p < 0.01) were higher, whereas the entropy was lower (p < 0.01). This method might be applicable to quantification of PRM damage within the muscle.

Development and validation of a composite AI model for the diagnosis of levator ani muscle avulsion

OBJECTIVE

To assess the feasibility and reliability of a composite AI model for the diagnosis of levator ani muscle (LAM) avulsion of tomographic ultrasound imaging (TUI).

METHODS

Ultrasonic images of the pelvic floor from a total of 304 patients taken from January 2018 to October 2020 were included. All patients included underwent standardized interviews and transperineal ultrasound (TPUS). Transfer-learning and ensemble-learning methods were adopted to develop the proposed model on the basis of three classic convolutional neural networks (CNN). Confusion matrix (CM) and the ROC statistic were used to assess the effectiveness of the proposed model. Gradient-weighted class activation mappings (Grad-CAMs) were used to help enhance the interpretability of the proposed model.

RESULTS

Of the 304 patients included, 208 were in the derivation cohort (108 LAM avulsion and 100 normal) and 96 (39 LAM avulsion and 57 normal) were in the validation cohort. The proposed model in LAM avulsion diagnosis outperformed other models and a junior clinician in both the test set of derivation cohort and the validation cohort, with accuracies of 0.95 and 0.81, and AUCs of 0.98 and 0.86, respectively. According to the heatmap of Grad-CAMs, the proposed model mainly localizes areas between the pubic symphysis and the bilateral insertion point of LAM when making a diagnosis, which is exactly the region of interest in clinical practice.

CONCLUSION

The proposed model using ultrasonic images of the pelvic floor may be a promising tool in assisting the diagnosis of LAM avulsion in clinical practice.

KEY POINTS

• First AI-assisted model for levator ani muscle avulsion diagnosis • Diagnosis accuracy of less-experienced clinicians could be improved using the proposed model.

The effect of vaginal delivery and Caesarean section on the anal Sphincter complex of Primipara based on optimized three-dimensional ultrasound image and nuclear regression Reconstruction Algorithm

Objective

The study used the optimized nuclear regression reconstruction algorithm to explore the value of three-dimensional perineal ultrasound evaluation of the effect of caesarean delivery and caesarean section on the anal sphincter complex of primipara.

Methods

This study performed three-dimensional perineal ultrasound scanning of the anal sphincter complex of 157 primiparas 42 days after delivery. Among them, 77 were in caesarean delivery (spontaneous delivery group) and 80 were in caesarean section (caesarean delivery group) from September 2018 to December 2020 in our hospital. The thickness of the end plane, the middle plane, the distal plane and the distal plane of the external anal sphincter at 3, 6, 9, 12 o'clock direction, and measure the thickness of the central plane of the pubic rectum muscle at 4, 8 o'clock direction. At the same time, the study used tomography and volume contrast imaging to observe the morphology and integrity of the anal sphincter complex.

Results

The thickness of the distal anal sphincter at the 12 o'clock direction, the proximal anal sphincter at 6, 12 o'clock, and the central plane at 9 and 12 o'clock in the obstetric group were smaller than those in the caesarean section group (all P < 0.05). There were no significant differences in the thickness of the remaining anal internal and external anal sphincter and puborectalis muscles between the two groups in different directions (all P> 0.05). In the obstetric group, a perineal sphincter defect was found via three-dimensional perineal ultrasound.

Conclusion

The delivery method has a certain influence on the shape of the anal sphincter complex. The thickness of the internal and external anal sphincter of the primiparous women in a certain direction is significantly smaller than that of caesarean section. Transperineally three-dimensional ultrasound can clearly show the morphological characteristics and integrity of the anal sphincter complex, and diagnose the defect of the anal sphincter complex.

Appearance of the levator ani muscle subdivisions on 3D transperineal ultrasound

BACKGROUND

The levator ani muscle (LAM) consists of different subdivisions, which play a specific role in the pelvic floor mechanics. The aim of this study is to identify and describe the appearance of these subdivisions on 3-Dimensional (3D) transperineal ultrasound (TPUS). To do so, a study designed in three phases was performed in which twenty 3D TPUS scans of vaginally nulliparous women were assessed. The first phase was aimed at getting acquainted with the anatomy of the LAM subdivisions and its appearance on TPUS: relevant literature was consulted, and the TPUS scan of one patient was analyzed to identify the puborectal, iliococcygeal, puboperineal, pubovaginal, and puboanal muscle. In the second phase, the five LAM subdivisions and the pubic bone and external sphincter, used as reference structures, were manually segmented in volume data obtained from five nulliparous women at rest. In the third phase, intra- and inter-observer reproducibility were assessed on twenty TPUS scans by measuring the Dice Similarity Index (DSI).

RESULTS

The mean inter-observer and median intra-observer DSI values (with interquartile range) were: puborectal 0.83 (0.13)/0.83 (0.10), puboanal 0.70 (0.16)/0.79 (0.09), iliococcygeal 0.73 (0.14)/0.79 (0.10), puboperineal 0.63 (0.25)/0.75 (0.22), pubovaginal muscle 0.62 (0.22)/0.71 (0.16), and the external sphincter 0.81 (0.12)/0.89 (0.03).

CONCLUSION

Our results show that the LAM subdivisions of nulliparous women can be reproducibly identified on 3D TPUS data.

3D Ultrasound Strain Imaging of Puborectalis Muscle

The female pelvic floor (PF) muscles provide support to the pelvic organs. During delivery, some of these muscles have to stretch up to three times their original length to allow passage of the baby, leading frequently to damage and consequently later-life PF dysfunction (PFD). Three-dimensional (3D) ultrasound (US) imaging can be used to image these muscles and to diagnose the damage by assessing quantitative, geometric and functional information of the muscles through strain imaging. In this study we developed 3D US strain imaging of the PF muscles and explored its application to the puborectalis muscle (PRM), which is one of the major PF muscles.

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.