Deep learning-based pelvic levator hiatus segmentation from ultrasound images

Huang, Zeping; Qu, Enze; Meng, Yi-Shuang; Zhang, Man; Wei, Qiuwen; Bai, Xiang-Hui; Zhang, Xinling

doi:10.1016/j.ejro.2022.100412

Cited by 6 publications

(3 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Therefore, we cannot compare our results with those of other authors. Previous authors have applied DL to view other structures of the pelvic floor on ultrasound, with disparate ISD results [4][5][6]15]. Studies that examined the hiatus area of the elevator using DL reported DSIs greater than 0.9 [4,5,15,16], establishing a value of 0.94 in the case of the urogenital hiatus [17].…”

Section: Discussionmentioning

confidence: 99%

“…Previous authors have applied DL to view other structures of the pelvic floor on ultrasound, with disparate ISD results [4][5][6]15]. Studies that examined the hiatus area of the elevator using DL reported DSIs greater than 0.9 [4,5,15,16], establishing a value of 0.94 in the case of the urogenital hiatus [17]. However, when using the DL to study solid structures, as in the case of the levator ani muscle, the DSI is lower, ranging between 0.6 and 0.77 [4,18,19].…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

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

García-Mejido,

Solis-Martín,

Martín-Morán

et al. 2024

Int Urogynecol J

View full text Add to dashboard Cite

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.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

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

García-Mejido,

Solis-Martín,

Martín-Morán

et al. 2024

Int Urogynecol J

View full text Add to dashboard Cite

show abstract

“…The latter model also had a good segmentation effect (ICC = 0.91-0.96). Huang [13] et al also verified the automatic segmentation model on different instrument images and achieved good results (DSC = 0.952-0.964). The above models constitute progress in the segmentation of LAM defects, but they are still based on reconstructed LAM images.…”

Section: Fully Automatic Segmentation Model Based On Reconstructed Le...mentioning

confidence: 92%

Advanced Application of Artificial Intelligence for Pelvic Floor Ultrasound in Diagnosis and Treatment

Qu¹,

Zhang²

2023

Advanced Ultrasound in Diagnosis and Therapy

View full text Add to dashboard Cite

Artificial intelligence-based pelvic floor ultrasound helps the diagnosis, preoperative assessment, and postoperative monitoring of female pelvic floor dysfunction (FPFD). The application of artificial intelligence in pelvic floor ultrasound mainly includes automatic segmentation and measurement, the diagnosis of muscle injury, childbirth prediction and postoperational evaluation. It can not only overcome the problem of operator experience dependence but also improve work efficiency and simplify the workflow, which has popularized the application of pelvic floor ultrasound. However, most of the current research is still limited to the automatic segmentation of three-dimensional axial plane levator hiatus (LH). The automatic reconstruction, real-time tracking of 3D/4D images and the imaging navigation of pelvic floor surgery remain major challenges for researchers.

show abstract

A new method for evaluating the uniformity of steel slag distribution in steel slag asphalt mixture based on deep learning

Liu

Yuan

Wang

et al. 2023

Construction and Building Materials

View full text Add to dashboard Cite

Deep learning-based pelvic levator hiatus segmentation from ultrasound images

Cited by 6 publications

References 24 publications

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

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

Advanced Application of Artificial Intelligence for Pelvic Floor Ultrasound in Diagnosis and Treatment

A new method for evaluating the uniformity of steel slag distribution in steel slag asphalt mixture based on deep learning

Contact Info

Product

Resources

About