Interpretable machine learning for thyroid cancer recurrence predicton: Leveraging XGBoost and SHAP analysis

•Train and test an XGBoost model for recurrence prediction of patients suffering from thyroid cancer with high accuracy.•Use SHAP values to identify relevant biomarkers.•Use SHAP dependence plots to identify threshold values for patients at risk.•Our results could improve the identification of patie...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:European journal of radiology Jg. 186; S. 112049
Hauptverfasser: Schindele, Andreas, Krebold, Anne, Heiß, Ursula, Nimptsch, Kerstin, Pfaehler, Elisabeth, Berr, Christina, Bundschuh, Ralph A., Wendler, Thomas, Kertels, Olivia, Tran-Gia, Johannes, Pfob, Christian H., Lapa, Constantin
Format: Journal Article
Sprache:Englisch
Veröffentlicht: Ireland Elsevier B.V 01.05.2025
Schlagworte:
Adult
Aged
Biomarkers, Tumor - blood
Boosting Machine Learning Algorithms
Female
Germany - epidemiology
Humans
Machine Learning
Male
Middle Aged
Neoplasm Recurrence, Local - diagnosis
Neoplasm Recurrence, Local - epidemiology
Prognosis
Reproducibility of Results
Retrospective Studies
Risk Assessment
Risk Factors
Sensitivity and Specificity
Thyroid Neoplasms - diagnosis
Thyroid Neoplasms - diagnostic imaging
Thyroid Neoplasms - epidemiology
Germany
ISSN:0720-048X, 1872-7727, 1872-7727
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:•Train and test an XGBoost model for recurrence prediction of patients suffering from thyroid cancer with high accuracy.•Use SHAP values to identify relevant biomarkers.•Use SHAP dependence plots to identify threshold values for patients at risk.•Our results could improve the identification of patients with high risk of tumor recurrence. For patients suffering from differentiated thyroid cancer (DTC), several clinical, laboratory, and pathological features (including patient age, tumor size, extrathyroidal extension, or serum thyroglobulin levels) are currently used to identify recurrence risk. Validation and potential adjustment of their individual and combined prognostic values using a large patient cohort with several years of follow-up might improve the correct identification of patients at risk. In this retrospective study, we developed an XGBoost model using clinical and biomarker features for accurate DTC recurrence prediction using a cohort of 1228 consecutive patients (965 papillary, and 263 follicular) that were treated at the Department of Nuclear Medicine at University Hospital Augsburg between 1976 and 2010. The dataset was split into 982 patients for model training, and 246 for independent testing. From the 982 patients, 200 different random combinations of 785 training and 197 validation patients were conducted. To identify critical risk factors and understand the model’s decision-making process, we conducted Shapely Additive exPlanations (SHAP) analysis. The XGBoost model achieved an AUROC of 0.84 (95 % CI: 0.84–0.86; SD: 0.08), sensitivity of 0.79 (95 % CI: 0.77–0.81; SD: 0.17), and specificity of 0.78 (95 % CI: 0.77–0.79; SD: 0.04) on the validation datasets, and an AUROC of 0.88 (sensitivity 0.83, specificity 0.80) on the independent test set. Tumor size, maximal thyroglobulin values within six months after thyroidectomy, and maximal thyroglobulin antibody levels within 12 to 24 months after thyroidectomy were the most important factors. SHAP dependence plots suggested new recurrence risk thresholds for a tumor size of 25 mm, maximal serum thyroglobulin levels of 3 and 10 ng/mL, respectively, and maximal thyroglobulin antibody levels of 120 IU/mL. Our XGBoost model, supported by SHAP analysis empowers clinicians with interpretable insights and defined risk thresholds and could facilitate informed decision-making and patient-centric care.
Bibliographie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
ISSN:0720-048X
1872-7727
1872-7727
DOI:10.1016/j.ejrad.2025.112049