Studies on predicting the outcome of sling surgery are limited. Most depend on analysis of multiple confounding factors using regression models. However, their prediction results are limited. In this study, we tested a statistical regression model and an AI model for the prediction of the outcome of mid-urethral sling.

All women underwent MUS in our facility from January 2002 to January 2020 with a minimum follow- up of 1 year, were retrospectively studied. All methods were performed in accordance with the relevant guidelines and regulations.

Inclusion and exclusion criteria were similar to previous report 7. 257 patients were contacted by phone and asked to attend an outpatient clinic visit. Informed consent was obtained from all patients. The study design and protocol were approved by the local ethical/scientific committee of UNC.

Confounding factors were: age, body mass index (BMI), parity, previous pelvic surgery, pre-operative urodynamics (UDS).The follow up visits, included per vaginal examination, stress test, pad test, post-void residual (PVR) and symptom scores.

The primary outcome is the construction of a prediction model that selects the patient with the best success rate. Cure is defined according to objective criteria (a negative stress test, a negative 1-hour pad test and no retreatment) and subjective criteria (self-reported absence of symptoms, no leakage episodes). Failure was defined as persistent stress component.

Patients’ data were retrieved and reviewed regarding demographic data, surgical history, preoperative examination, pre-operative UDS, type of the sling, concomitant repair of prolapse, and BMI which was classified into values of more or less than 30. Parity was classified into values of more or less than 3. Abdominal leak point pressure (ALPP) was classified into 3 grades; > 90 (grade 1), 90 − 60 (grade 2) and < 60 (grade 3)14. Pre-operative bladder capacity was classified as normal or low (< 250 ml). The type of sling used was TVT, TOT or PVS.

The study was divided into two phases. Phase I included the construction of a prediction model using binomial logistic regression (LR). In phase II, we applied an AI preferences (Support Vector Machines (SVM) and Artificial neural network (ANN) trying to obtain better predictions.

Phase I: The logistic regression model predicted the outcome of surgery with overall accuracy of 90.7% and positive predictive value of 61.5% [X2 (11) = 46.24, P < 0.001].

Phase II: The data of the patients were entered as 10 features; 9 were predictors and the 10th was the output. The output comprised 18 cases designated as ‘failure’ and 133 as ‘success’ output. The best model performance-wise was the (SVM) with 92% accuracy and 96% F1-score, which meets the industrial standards for predictive models. However, ANN produced 90% accuracy and 94% F1-score. However, our sample size is small

Building a statistical model regarding outcome of MUS surgery with high accuracy and sensitivity is applicable through LR. The use of AI is a good alternative to obtain valuable prediction of outcome of sling surgery. A larger sample size is needed to obtain better prediction. We plan to further include more cases to our model so that we can improve predictive outcome.

Prediction of the outcome of MUS surgery was achieved using different modalities with the best prediction of the outcome obtained by SVM method.

Ebbesen MH, Hunskaar S, Rortveit G et al: Prevalence, incidence and remission of urinary incontinence ?in women: longitudinal data from the Norwegian HUNT study (EPINCONT). BMC urol. 2013; 13(1):1–10.?
Dreiseitl S, Ohno-Machado L: Logistic regression and artificial neural network classification models: a ?methodology review. J Biomed Inform. 2002; 35(5):352-9.?
Jelovsek JE, Hill AJ, Chagin KM, et al: Predicting Risk of Urinary Incontinence and Adverse Events After ?Midurethral Sling Surgery in Women. Obstet.Gynecol. 2016; 127(2).?