Subtype-Dependent Performance of Cox and Machine Learning Survival Models for Recurrence Prediction in Breast Cancer: Development and External Validation Using Public Clinical Data

Francis Mawutor Amuyao ()

International Journal of Innovative Science and Research Technology (IJISRT), 2026, vol. 11, issue 06, 3298-3304

Abstract: Breast cancer recurrence risk prediction informs adjuvant treatment decisions and follow-up planning. Molecular subtypes capture biologically distinct risk profiles, yet whether machine learning (ML) survival methods offer consistent advantages over Cox proportional hazards (PH) modelling across subtypes remains unclear. We analysed 1,964 breast cancer patients across five molecular subtypes. Penalised Cox PH, Random Survival Forest (RSF), and Gradient Boosting Survival (GBS) models were developed for recurrence-free survival (RFS) prediction using discrimination, calibration, decision curve analysis, and subtype-stratified SHAP explainability.

Keywords: Breast Cancer; Recurrence Prediction; Survival Analysis; Molecular Subtype; Random Survival Forest; SHAP; METABRIC; Cox Proportional Hazards; External Validation. (search for similar items in EconPapers)
Date: 2026
References: Add references at CitEc
Citations:

Downloads: (external link)
https://www.ijisrt.com/subtypedependent-performanc ... public-clinical-data (application/pdf)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:cvr:ijisrt:2026:06:ijisrt26jun1500

DOI: 10.38124/ijisrt/26jun1500

Access Statistics for this article

More articles in International Journal of Innovative Science and Research Technology (IJISRT) from IJISRT Publication
Bibliographic data for series maintained by Rahul Goyel ().