Interpretable XGBoost model identifies idiopathic central precocious puberty in girls using four clinical and imaging features

Lu Tian ¹, Yan Zeng ², Helin Zheng ¹

⁰Department of Radiology, National Clinical Research Center for Child Health and Disorders, Ministry of Education Key Laboratory of Child Development and Disorders, Chongqing Key Laboratory of Pediatric Metabolism and Inflammatory Diseases, Children's Hospital of Chongqing Medical University, Chongqing, 400014, China.

Bmc Endocrine Disorders +

|

July 2, 2025

View abstract on PubMed

Summary

Machine learning models can now help diagnose idiopathic central precocious puberty (ICPP) in girls using key clinical data, potentially replacing the need for the costly gonadotropin-releasing hormone (GnRH) stimulation test.

Area Of Science

Pediatric Endocrinology
Medical Informatics
Machine Learning in Healthcare

Background

Idiopathic central precocious puberty (ICPP) diagnosis currently relies on the gonadotropin-releasing hormone (GnRH) stimulation test.
This gold standard test is expensive and time-consuming.
There is a need for alternative diagnostic methods.

Purpose Of The Study

To develop interpretable machine learning (ML) models for ICPP identification in girls.
To avoid the necessity of the GnRH stimulation test.
To provide a more accessible diagnostic tool.

Main Methods

246 female pediatric patients with early puberty were analyzed.
The least absolute shrinkage and selection operator (LASSO) identified key parameters: uterine volume, bone age/chronological age (BA/CA), basal FSH, and basal LH.
Logistic regression and five ML models (SVM, GaussianNB, XGBoost, RF, kNN) were constructed and evaluated using AUROC, accuracy, sensitivity, and specificity.

Main Results

Four key parameters (uterine volume, BA/CA, basal FSH, basal LH) were selected for ICPP diagnosis.
ML models achieved AUCs from 0.65 to 0.90 in the validation set.
The XGBoost model showed superior performance with the highest AUC, accuracy, specificity, and sensitivity, demonstrating excellent calibration and clinical utility.

Conclusions

An accurate and interpretable ML-based model was developed for ICPP diagnosis.
This model can assist clinicians in decision-making.
The developed model offers a potential alternative to the GnRH stimulation test.

Keywords:

Girl Idiopathic central precocious puberty Interpretable Machine learning

Related Concept Videos

Signs of Puberty

Puberty is a critical phase, typically beginning between the ages of 8 and 13 in girls and 9 and 14 in boys, though timing can vary based on genetics, environmental factors, and overall health. This period is characterized by the development of secondary sexual characteristics and the attainment of reproductive potential. Endocrine changes underpin puberty, with hormonal surges of Luteinizing Hormone (LH) and Follicle-Stimulating Hormone (FSH) instigated by Gonadotropin-Releasing Hormone (GnRH)...