Development and verification of machine learning model based on anogenital distance, penoscrotal distance, and 2D:4D finger ratio before puberty to predict hypospadias classification.

Objectives: To describe the anatomical abnormalities of hypospadias before puberty using current commonly used anthropometric index data and predict postoperative diagnostic classification. Methods: Children with hypospadias before puberty who were initially treated at Sichuan Provincial People's Hospital from April 2021 to September 2022 were selected. We recorded their preoperative penoscrotal distance, anogenital distance, 2D:4D finger ratio, and postoperative hypospadias classification. The receiver operating character curve was used for univariate analysis of the diagnostic predictive value of each index for hypospadias classification in the training set. Binary logistic regression, random forest, and support vector machine models were constructed. In addition, we also prospectively collected data from October 2022 to September 2023 as a test set to verify the constructed machine learning models. Results: This study included 389 cases, with 50 distal, 167 midshaft, and 172 proximal cases. In the validation set, the sensitivity of the binary LR, RF, and SVM was 17%, 17% and 0% for identifying the distal type, 61%, 55% and 64% for identifying the midshaft type, and 56%, 60% and 48% for identifying the proximal type, respectively. The sensitivity of the three-classification RF and SVM models was 17% and 17% for distal type, 64% and 73% for midshaft type, 60% and 60% for proximal type, respectively. In the Testing set, the sensitivity of the binary LR, RF and SVM was 6%, 0% and 0% for identifying the distal type, 64%, 55% and 66% for identifying the midshaft type, and 48%, 62% and 39% for identifying the proximal type, respectively. The sensitivity of the three-classification RF and SVM models was 12% and 0% for distal type, 57% and 77% for midshaft type, and 65% and 53% for proximal type, respectively. Compared with binary classification models, the sensitivity of the three-classification models for distal type was not improved. Conclusion: Anogenital distance and penoscrotal distance have a favorable predictive value for midshaft and proximal hypospadias, among which AGD2, with higher test efficiency and stability, is recommended as the preferred anogenital distance indicator. The 2D:4D finger ratio (RadioL, RadioR) has little predictive value for hypospadias classification.