Development and validation of diagnostic prediction models for central precocious puberty in girls based on machine learning: a multicenter retrospective study - Takeaways - MDSpire

Development and validation of diagnostic prediction models for central precocious puberty in girls based on machine learning: a multicenter retrospective study

  • By

  • Wenyong Wu

  • Zhe Su

  • Haiyan Wei

  • Yanhong Li

  • Benlong Zhu

  • Xin Yuan

  • Daibin Lei

  • Yi Wei

  • Xian Wu

  • Hanghan Ou

  • Xinyu Chen

  • Ziling Zhu

  • Ruimin Chen

  • June 15, 2026

  • 0 min

Share

  • 1

    Novel diagnostic prediction models for central precocious puberty (CPP) in girls were developed using machine learning techniques.

  • 2

    The study included 2148 girls with precocious puberty from four centers in China, divided into training, validation, and test groups.

  • 3

    Eight independent predictors for CPP diagnosis were identified, including chronological age and basal luteinizing hormone levels.

  • 4

    The support vector machine (SVM) model showed optimal performance with AUC values of 0.850 and 0.827 in internal and external validations, respectively.

  • 5

    The SVM prediction model demonstrated good discrimination and calibration for diagnosing CPP without the need for a GnRH stimulation test.

Original Source(s)

Development and validation of diagnostic prediction models for central precocious puberty in girls based on machine learning: a multicenter retrospective study

  • by Wenyong Wu, Zhe Su, Haiyan Wei, Yanhong Li, Benlong Zhu, Xin Yuan, Daibin Lei, Yi Wei, Xian Wu, Hanghan Ou, Xinyu Chen, Ziling Zhu, Ruimin Chen

  • June 15, 2026

Related Content

A 10-Year Cost-Effectiveness Analysis of Metabolic and Bariatric Surgery in Adolescents

Menopause in the spotlight: important research and clinical advances

  • June 16, 2026

Development and validation of a clinical prediction model for poor ovarian response in assisted reproductive technology