To evaluate the feasibility of Full-stack Smart Pelvic Floor Ultrasound (FSPFU) software in the acquisition and measurement of the minimal levator hiatus (LH).
Methods:
Transperineal pelvic floorultrasonography was performed in 119 women of 6-month postpartum from Nov.2020 to Jan.2021 of Shenzhen Second People′s Hospital. Mid-sagittal plane of pelvic floor was set as the initial plane, and the three-dimensional volume data was acquired. The dataset was stored in the machine. The offline volume data was manually adjusted to obtain the minimal LH images and measured by four physicians (two junior physicians as the D1 group and two senior physicians as the D2 group). For comparison, the results were also obtained using the fully automated method—the FSPFU software by a junior physician (the D3 group). The obtained parameters of minimal LH included area, circumference, anterioposterior diameter, transverse diameter, left and right levator-urethral gap distance. Analysistime was recorded for each group. The contours of minimal LH were outlined by three groups and the overlapping rate was calculated. The quality of the resulted images was evaluated and scored by another two senior physicians(A and B) independently.
Results:
The D3 group had a significant shorter analysistime compared with the other two groups, and the D1 group took a longer time than the D2 group, regardless of the cystocele severity (D1 82.97 s, D2 62.51 s, D3 2.71 s, all P<0.05). The intergroup agreements and correlations of the minimum LH area were good (all ICC>0.85, rs>0.70, P<0.001) and the outlined contours were largely overlapped (>92%). There was no significant difference in image quality among the three groups(all P>0.05).