Objective:To explore the clinical application value of convolutional neural network（CNN）based on deep learning in the automatic measurement of dynamic pelvic floor ultrasound video parameters and the diagnosis and classification of cystocele.Methods:A retrospective analysis was conducted on dynamic pelvic floor ultrasound videos from 398 postpartum women who underwent examinations at the First People's Hospital of Foshan between June 2020 and June 2022. The lowest point of the posterior bladder wall（PWB），urethral rotation angle（URA），and retrovesical angle（RVA）were manually measured by a senior radiologist（R1）and a junior radiologist（R2），and cystocele was classified according to the Green standard. The CNN model was employed to automatically extract the above parameters and to diagnose and classify cystocele. Using R1 measurements as a reference，intraclass correlation coefficient（ICC）was used to evaluate the consistency between the CNN model and R1，as well as between R2 and R1. The Kappa value was used to assess the agreement between the CNN model，R2，and R1 in the diagnosis and classification of cystocele. Additionally，the time consumption of the three measurement methods was compared.Results:The CNN model showed good consistency with R1 in measuring PWB and URA（ICC = 0.983，0.894），while its consistency in measuring RVA was moderate（ICC = 0.614）. The ICC between R2 and R1 in measuring PWB，URA，and RVA was 0.979，0.815，and 0.627，respectively. In the measurement of PWB and URA，the consistency between the CNN model and R1 was superior to that between R2 and R1. For cystocele diagnosis，the Kappa value between the CNN model and R1 was 0.924，which was higher than that between R2 and R1（0.904）. In cystocele classification，the Kappa value between the CNN model and R1 was 0.503，also higher than that between R2 and R1（0.426）. The CNN model processed a single video in 2.5（0.6）s，significantly faster than R1［59.9（16.9）s］and R2［56.8（11.2）s］（all P < 0.001）. Conclusions:The CNN model demonstrates high accuracy and efficiency in the measurement，diagnosis，and classification of cystocele，outperforming a junior radiologist and showing potential for clinical application.

Objective:To explore the clinical application value of convolutional neural network（CNN）based on deep learning in the automatic measurement of dynamic pelvic floor ultrasound video parameters and the diagnosis and classification of cystocele.Methods:A retrospective analysis was conducted on dynamic pelvic floor ultrasound videos from 398 postpartum women who underwent examinations at the First People's Hospital of Foshan between June 2020 and June 2022. The lowest point of the posterior bladder wall（PWB），urethral rotation angle（URA），and retrovesical angle（RVA）were manually measured by a senior radiologist（R1）and a junior radiologist（R2），and cystocele was classified according to the Green standard. The CNN model was employed to automatically extract the above parameters and to diagnose and classify cystocele. Using R1 measurements as a reference，intraclass correlation coefficient（ICC）was used to evaluate the consistency between the CNN model and R1，as well as between R2 and R1. The Kappa value was used to assess the agreement between the CNN model，R2，and R1 in the diagnosis and classification of cystocele. Additionally，the time consumption of the three measurement methods was compared.Results:The CNN model showed good consistency with R1 in measuring PWB and URA（ICC = 0.983，0.894），while its consistency in measuring RVA was moderate（ICC = 0.614）. The ICC between R2 and R1 in measuring PWB，URA，and RVA was 0.979，0.815，and 0.627，respectively. In the measurement of PWB and URA，the consistency between the CNN model and R1 was superior to that between R2 and R1. For cystocele diagnosis，the Kappa value between the CNN model and R1 was 0.924，which was higher than that between R2 and R1（0.904）. In cystocele classification，the Kappa value between the CNN model and R1 was 0.503，also higher than that between R2 and R1（0.426）. The CNN model processed a single video in 2.5（0.6）s，significantly faster than R1［59.9（16.9）s］and R2［56.8（11.2）s］（all P < 0.001）. Conclusions:The CNN model demonstrates high accuracy and efficiency in the measurement，diagnosis，and classification of cystocele，outperforming a junior radiologist and showing potential for clinical application.

Objective:To investigate the significance of clinical factors combined with transvaginal ultrasound and contrast-enhanced ultrasound(CEUS) in guiding the choice of treatment plan for cesarean scar pregnancy(CSP).Methods:The clinical and transvaginal ultrasound and CEUS data of 120 patients with CSP from January 2016 to June 2021 in the First People′s Hospital of Foshan were retrospectively analyzed, and they were divided into ultrasound-guided curettage/ hysteroscopic group (Group A, 91 cases) and laparoscopic group (Group B, 29 cases) according to treatment option, and the differences in clinical and ultrasound factors between the two groups were compared, and to determine the relevant clinical and ultrasound indicators for the choice of treatment option.Results:There were statistical differences between the 2 groups in comparison of whether the gestational sac/mass protruded toward the plasma membrane, gestational sac/mass diameter, the main blood supply site of the gestational sac/mass, the site of the chorion/early placenta and scar thickness (all P<0.05). Logistic regression analysis indicated that CEUS showing major blood supply site of the gestational sac/mass ( OR=6.029, P=0.003) and uterine scar thickness ( OR=12.998, P=0.002) were independent risk factors for minimally invasive surgery for CSP. Conclusions:Ultrasound combined with clinical factors have a certain value in the selection of treatment options for CPS, and the thickness of the uterine scar and the main blood supply site of the gestational sac/mass showed in CEUS may be key factors affecting the minimally invasive surgical treatment of CSP.

Objective To investigate the feasibility and accuracy of transperineal real‐time three‐dimensional ultrasound combined with clinical factors in predicting the risk of female stress urinary incontinence( SUI ) . Methods T hree hundred and forty‐eight female patients with SUI diagnosed were selected as the case group ,and 102 healthy people in the same period were selected as the control group . All subjects underwent transperineal real‐time three‐dimensional ultrasound . T he ultrasonic parameters of resting state ,contraction and Valsalva were measured ,and the clinical parameters such as age ,height , weight ,history of pregnancy and childbirth were collected . According to the time sequence ,all the subjcets were divided into derivation cohort and verification cohort inproportion to 2∶1 ,single factor screening and logistic multiple regression analysis were carried out on 24 factors ,and the risk model was established . T he cut‐off value of the disease probability P was determined by the ROC curve of the subjects ,and then the accuracy of the cut‐off value in predicting SUI was verified in the verification group . Results Single factor analysis showed that 13 parameters were associated with SUI( all P ＜0 .05) . Logit P＝2 .014+1 .870× Z1 was established by multivariate logistic regression analysis . T he cut‐off value of the disease probability P determined by ROC curve was 0 .823 . T he predictive sensitivity of the model was 68 .1% ( 95% CI : 59 .6% －76 .6% ) ,specificity was 91 .2% ( 95% CI :86 .0% －96 .4% ) ,positive predictive value was 64 .3%( 95% CI : 55 .6% － 73 .0% ) and negative predictive value was 92 .5% ( 95% CI : 86 .2% － 98 .8% ) . Conclusions It is feasible to predict the risk of female stress urinary incontinence by transperineal real‐time three‐dimensional ultrasound combined with clinical factors . Although ,some limitations with the prediction model ,it has accuracy in predicting SUI with obvious symptoms .

Objective: To investigate the feasibility and accuracy of transperineal real-time three-dimensional ultrasound combined with clinical factors in predicting the risk of female stress urinary incontinence(SUI). Methods: Three hundred and forty-eight female patients with SUI diagnosed were selected as the case group, and 102 healthy people in the same period were selected as the control group. All subjects underwent transperineal real-time three-dimensional ultrasound. The ultrasonic parameters of resting state, contraction and Valsalva were measured, and the clinical parameters such as age, height, weight, history of pregnancy and childbirth were collected. According to the time sequence, all the subjcets were divided into derivation cohort and verification cohort inproportion to 2∶1, single factor screening and logistic multiple regression analysis were carried out on 24 factors, and the risk model was established. The cut-off value of the disease probability P was determined by the ROC curve of the subjects, and then the accuracy of the cut-off value in predicting SUI was verified in the verification group. Results: Single factor analysis showed that 13 parameters were associated with SUI(all P<0.05). Logit P=2.014+ 1.870×Z₁ was established by multivariate logistic regression analysis. The cut-off value of the disease probability P determined by ROC curve was 0.823. The predictive sensitivity of the model was 68.1% (95%CI: 59.6%-76.6%), specificity was 91.2% (95%CI: 86.0%-96.4%), positive predictive value was 64.3% (95%CI: 55.6%-73.0%) and negative predictive value was 92.5% (95%CI: 86.2%-98.8%). Conclusions It is feasible to predict the risk of female stress urinary incontinence by transperineal real-time three-dimensional ultrasound combined with clinical factors. Although, some limitations with the prediction model, it has accuracy in predicting SUI with obvious symptoms.