I am an endocrinologist at the Pituitary Tumor Center, Department of Neurosurgery, The Sixth Affiliated Hospital of Sun Yat-sen University. My clinical work focuses on pituitary and neuroendocrine disorders, while my research spans diabetes heterogeneity and the translation of computational methods into clinically meaningful patient stratification.

My work connects clinical medicine with:

• Pituitary and neuroendocrine medicine — multidisciplinary evaluation of pituitary tumors and related endocrine disorders
• Diabetes phenotyping and classification — identifying meaningful subgroups beyond conventional diagnostic labels
• Clinical machine learning — developing and evaluating models with attention to discrimination, calibration, transportability, and clinical utility
• Bioinformatics — investigating disease mechanisms and molecular heterogeneity
• Clinical epidemiology — cohort studies, survival analysis, risk modelling, and robust sensitivity analyses
• Responsible medical AI — evaluating AI systems under real-world clinical constraints

• Real-world performance of open-source large language models in diabetes diagnosis
  Frontiers in Endocrinology, 2026 · Article

• A new vision of diabetes classification: A shift from clinical manifestation to etiological molecular mechanism
  Journal of Practical Medicine, 2024 · DOI

• Application of the Chinese Expert Consensus on Diabetes Classification in Clinical Practice
  Chinese Journal of Internal Medicine, 2023 · PubMed

• Validation of the Swedish diabetes re-grouping scheme in adult-onset diabetes in China
  The Journal of Clinical Endocrinology & Metabolism, 2020 · PubMed

• Multidisciplinary research on pituitary tumors, endocrine phenotypes, and clinically relevant outcomes
• Reproducible workflows spanning study design, analysis, peer review, and publication
• Public-cohort and real-world-data research with explicit evidence and validation boundaries
• Medical AI evaluation that prioritizes clinical relevance over model complexity
• Research infrastructure for a focused, AI-assisted one-person lab

Clinical relevance before model complexity.
Reproducibility before novelty claims.
Transparent negative results and clear evidence boundaries.

I maintain personal forks of the One Person Lab workflow stack while adapting it to medical-research practice:

• one-person-lab — a federated gateway for a focused research lab
• opl-flow — reusable Codex workflows for research and project execution

• Academic profile: FengGao Lab @ SYSU
• Publications: Google Scholar
• Research identity: ORCID 0000-0003-2812-3873
• Email: yang_shting@163.com