-
1
Pelvic floor dysfunction (PFD) is common after total hysterectomy, necessitating efficient assessment methods due to its impact on quality of life.
-
2
The study proposes an automated analysis of transperineal ultrasound (TPUS) using deep learning to improve efficiency and objectivity in PFD evaluation.
-
3
A labeled dataset was created for anatomical landmarks, and a multi-scale shifted window Transformer was developed for automatic segmentation and key point detection.
-
4
The model achieved an average Dice coefficient of 88.67% in segmentation, with high consistency to manual annotations, indicating its reliability.
-
5
This automated method is applicable to both benign and gynecologic oncology patients, providing a valuable tool for postoperative monitoring and rehabilitation.
