Related Experiment Video
Updated: Jul 3, 2026
Clinical Application of 24 G Cannula Needle and 3-0 Polypropylene Suture in Vas Deferens Exploration
Published on: February 10, 2023
To explore or not: machine learning models for intraoperative decision on testicular exploration in infants under 3
Yonghua Fu1,2, Guobin Liu1,2, Zhendi Tang1,2
1National Clinical Research Center for Child Health and Disorders, Ministry of Education Key Laboratory of Child Development and Disorders, Chongqing Key Laboratory of Pediatrics, Chongqing, 400014, China.
Insights
A machine learning model accurately predicts testicular necrosis risk in infants with incarcerated inguinal hernia (IIH). This tool integrates ultrasound and lab data to guide surgical decisions and improve outcomes.
Area of Science:
- Pediatric Surgery
- Machine Learning in Medicine
- Diagnostic Imaging
Background:
- Incarcerated inguinal hernia (IIH) in infants poses a risk of testicular damage.
- Current methods for assessing testicular viability preoperatively are limited.
- Objective prediction of necrosis is crucial for timely intervention.
Purpose of the Study:
- Develop a machine learning (ML) model for preoperative prediction of testicular necrosis risk in male infants under 3 months with IIH.
- Address limitations of current assessment methods for testicular viability.
- Enhance clinical decision-making in IIH management.
Main Methods:
- Retrospective analysis of 288 male infants with IIH undergoing emergency surgery.
- Training ten ML models using preoperative variables: testicular blood flow, echotexture, incarceration duration, procalcitonin, and neutrophil-to-lymphocyte ratio.
- Evaluating model performance using ROC AUC, PR AUC, accuracy, precision, recall, F1-score, and SHAP analysis for interpretability.
Main Results:
- The Gradient Boosting ML model demonstrated superior performance with a ROC AUC of 0.940 and recall of 0.889.
- SHAP analysis identified key predictors: absent testicular blood flow, heterogeneous echotexture, prolonged incarceration, and elevated procalcitonin and neutrophil-to-lymphocyte ratio.
- The model provides an interpretable prediction of testicular necrosis risk.
Conclusions:
- An ML model effectively predicts preoperative testicular necrosis risk in infants with IIH.
- Integration of ultrasound, serological markers, and clinical data offers an interpretable tool.
- External validation in multi-center prospective cohorts is necessary before clinical implementation to optimize outcomes.
Purpose:
To develop a machine learning (ML) model for preoperative prediction of testicular necrosis risk in male infants under 3 months with incarcerated inguinal hernia (IIH), addressing the limitations of current assessment methods.
Methods:
We retrospectively analyzed 288 male infants under 3 months with IIH who underwent emergency surgery. Key preoperative variables (testicular blood flow, echotexture, incarceration duration, procalcitonin, neutrophil-to-lymphocyte ratio) were used to train ten ML models. Performance was evaluated using ROC AUC, PR AUC, accuracy, precision, recall, and F1-score. SHAP analysis assessed interpretability.
Results:
The Gradient Boosting model performed best, achieving a ROC AUC of 0.940 and a recall of 0.889. SHAP identified absent testicular blood flow, heterogeneous echotexture, prolonged incarceration, and elevated procalcitonin and neutrophil-to-lymphocyte ratio as top predictors.
Conclusion:
This ML model predicts testicular necrosis risk preoperatively. By integrating color Doppler ultrasound, serological markers, and clinical data, it offers an interpretable tool to guide selective testicular exploration, potentially optimizing outcomes. However, given the single-center retrospective nature of this study, external validation in multi-center prospective cohorts is required before clinical implementation.