Adult Circumcision Patient Outcomes Predictor

Disclaimer: Adult circumcision remains a safe and commonly performed procedure with many proven benefits, including reduced rates of urinary tract infection, penile cancer, and certain sexually transmitted infections. This tool is intended to help surgeons by providing an estimated probability of composite complications based on intraoperative vitals and patient characteristics. It should be used as an adjunct to, not a replacement for, clinical judgment and shared decision-making with the patient or guardian.

Full terms of use, limitations, and privacy

Intended use

CircumScore is intended for qualified surgeons and clinical staff. It uses measurements recorded during the procedure (blood loss, operative time, and intraoperative vitals), so it estimates complication risk for a procedure that has already been performed. Its intended role is postoperative: helping clinicians decide which patients may benefit from closer follow-up, earlier wound checks, adjusted analgesia, or more detailed discharge counseling. It should not be used for preoperative planning, technique selection, or patient self-assessment. Estimates are most reliable for patients similar to those in the development cohort (ages 18 to 93, blood loss 0 to 100 ml). Some combinations, such as laser circumcision in patients under 30 or traditional circumcision in patients over 80, were rare or absent, and estimates for them are less certain.

Research tool, not a medical device

This tool has not been cleared, approved, or certified by the U.S. Food and Drug Administration, the European Medicines Agency, or any other regulatory authority, and it is not CE-marked as a medical device. It is provided for research and educational purposes only.

Limitations

The underlying model was developed on 194 patients treated at a single center in Milan, Italy between 2023 and 2024, using retrospectively collected data. It has not undergone external validation, and its performance in other populations, care settings, or surgical practices is unknown. All estimates are probabilistic and may be inaccurate for any individual patient.

Clinical responsibility

Output from this tool is an adjunct to, and never a replacement for, clinical judgment, direct examination, and shared decision-making with the patient or guardian. The treating clinician remains solely responsible for all diagnostic and treatment decisions.

No warranty

This tool is provided “as is,” without warranty of any kind, express or implied, including but not limited to warranties of merchantability, fitness for a particular purpose, and accuracy. The authors and their affiliated institutions accept no liability for any loss, injury, or damage arising from use of, or reliance on, this tool.

Privacy

Values entered into the form are used only to compute the estimate shown. They are not stored by this application and are not shared with third parties.

Model Architecture & Performance

Published: Shpaner, L., Saitta, G., et al. BMC Urology (2026). doi:10.1186/s12894-026-02072-x

We compared three supervised models: logistic regression, random forest, and support vector machines to predict short-term complications after adult male circumcision. Data from 194 patients (≥18 years) at a Milan center (2023–2024), using age, BMI, blood loss, surgical technique (traditional vs. laser), intraop vitals, and comorbidities.

Resampling & Training

No resampling (SMOTE/ROS) was applied

Stratified 10-fold cross-validation

Standard preprocessing pipelines

Performance Metrics (SVM Best)

Metric	Point Estimate (95% CI)
AUC ROC	0.907 (0.855–0.950)
Precision / PPV	0.725 (0.623–0.826)
Average Precision	0.832 (0.735–0.913)
Sensitivity / Recall	0.862 (0.765–0.939)
Specificity	0.860 (0.797–0.915)
F1-Score	0.787 (0.706–0.857)
Brier Score	0.105 (0.077–0.134)
Model Threshold (Point Estimate)	0.238

Metric

Point Estimate (95% CI)

AUC ROC

0.907 (0.855–0.950)

Precision / PPV

0.725 (0.623–0.826)

Average Precision

0.832 (0.735–0.913)

Sensitivity / Recall

0.862 (0.765–0.939)

Specificity

0.860 (0.797–0.915)

F1-Score

0.787 (0.706–0.857)

Brier Score

0.105 (0.077–0.134)

Model Threshold (Point Estimate)

0.238

Top Predictors (SHAP)

Intraoperative blood loss and surgical technique were the strongest predictors.

Notes & Next Steps

Strengths: rigorous CV, interpretability via SHAP.
Limitations: single-center data, no external validation.
Future: test on diverse populations and settings.

Hyperparameter Tuning

Parameter	Options	Chosen
Kernel Type	linear, rbf, poly, sigmoid	rbf
Cost Parameter (C)	0.0001 – 100 (log scale)	100
Gamma (γ)	0.001, 0.01, 0.05, 0.1, 0.2, 0.5, scale, auto	auto
Resampling Method	None, SMOTE, ROS	None

Parameter

Options

Chosen

Kernel Type

linear, rbf, poly, sigmoid

rbf

Cost Parameter (C)

0.0001 – 100 (log scale)

100

Gamma (γ)

0.001, 0.01, 0.05, 0.1, 0.2, 0.5, scale, auto

auto

Resampling Method

None, SMOTE, ROS

None

Table 1. Baseline clinical characteristics by surgical technique

From Shpaner, L., Saitta, G. BMC Urology (2026). doi:10.1186/s12894-026-02072-x

Variable	Traditional Circumcision (n = 132)	Laser Circumcision (n = 62)	P-value
BMI	24.08 (3.03)	24.06 (2.97)	0.9665
Diastolic BP	86.97 (11.12)	74.35 (20.54)	< 0.001
Systolic BP	121.82 (11.37)	123.55 (7.26)	0.2126
Heart Rate (bpm)	76.48 (5.96)	74.76 (5.57)	0.0484
Pulse Oximetry (%)	96.70 (1.80)	96.66 (1.76)	0.8741
Intraoperative Blood Loss (ml)	11.59 (17.64)	0.03 (0.52)	< 0.001
Surgical Time (min)	28.70 (5.44)	27.06 (5.94)	0.0684
Age	–	–	< 0.001
Age 18–29	79 (59.85%)	6 (9.68%)	–
Age 30–39	16 (12.12%)	6 (9.68%)	–
Age 40–49	4 (3.03%)	6 (9.68%)	–
Age 50–59	4 (3.03%)	13 (20.97%)	–
Age 60–69	9 (6.82%)	21 (33.87%)	–
Age 70–79	10 (7.58%)	10 (16.13%)	–
Age 80–89	7 (5.30%)	0 (0.00%)	–
Age 90–99	3 (2.27%)	0 (0.00%)	–
Obese	8 (6.06%)	3 (4.84%)	0.9918
Overweight	34 (25.76%)	19 (30.65%)	0.5894
Diabetes	11 (8.33%)	19 (30.65%)	< 0.001
Bleeding, Edema, Pain, or Infection	57 (43.18%)	1 (1.61%)	< 0.001

Values are presented as mean (standard deviation) for continuous variables and number (percentage) for categorical variables. The ‘Underweight’ BMI category was evaluated but excluded from Table 1 due to low sample size (n = 3) and no significant between-group difference (p = 1.0). Bold values indicate statistically significant between-group differences (p < 0.001).