Interpretable machine learning for presurgical differentiation of Hürthle cell carcinoma and adenoma: a SHAP-augmented approach - Takeaways - MDSpire

Interpretable Machine Learning Techniques for Preoperative Distinction Between Hürthle Cell Carcinoma and Adenoma: A SHAP-Enhanced Methodology

  • By

  • Junyu Cao

  • Jing Li

  • Chuancheng Zhou

  • Jie Zhou

  • Kunxian Yang

  • July 20, 2026

Share

  • 1

    The study developed an interpretable machine learning framework to differentiate between Hürthle cell adenoma and carcinoma preoperatively.

  • 2

    A total of 554 patients were enrolled, with 280 diagnosed with HCA and 274 with HCC, using various clinical and serological variables.

  • 3

    The XGBoost model achieved an AUC of 0.911, significantly outperforming logistic regression, which had an AUC of 0.746.

  • 4

    SHAP analysis identified key predictors of malignancy, including vascularity grade, absent halo sign, and elevated serum thyroglobulin.

  • 5

    The SHAP-augmented XGBoost framework aims to enhance presurgical risk stratification and reduce unnecessary thyroidectomies.

Original Source(s)

Interpretable machine learning for presurgical differentiation of Hürthle cell carcinoma and adenoma: a SHAP-augmented approach

  • by Junyu Cao, Jing Li, Chuancheng Zhou, Jie Zhou, Kunxian Yang

  • July 20, 2026

Related Content

Global research landscape and hotspots of paraptosis: a multi-database bibliometric analysis with a focused literature review

Ovarian endometriosis associated with significantly elevated CA19-9: a case report

Prognostic significance of tumoral Siglec-15 expression in solid tumors: a systematic review and meta-analysis