A Unified Framework for Survival Prediction: Combining Machine Learning Feature Selection with Traditional Survival Analysis in Heart Failure and METABRIC Breast Cancer
Ein einheitlicher Rahmen zur Überlebensprognose: Kombination von Feature-Selektion mit Machine Learning und traditioneller Überlebensanalyse bei Herzinsuffizienz und METABRIC-Brustkrebs
Diagnostics · 06.03.2026 · 16 (5) · 790
DOI: 10.3390/diagnostics16050790 ↗Auf einen Blick
Die Studie kombinierte ML-gestützte Feature-Selektion mit Cox- und Kaplan–Meier-Analyse. In beiden Kohorten zeigte der Rahmen eine stabile Risikostratifizierung mit klaren Überlebensunterschieden.
Fragestellung
Kann eine Kombination aus ML-basierter Feature-Selektion und klassischer Überlebensanalyse robuste und klinisch interpretierbare Prognosegruppen erzeugen?
Methode
ML-Feature-Selektion mit GBM-Cox, RSF und LASSO-Cox; 5-fold stratified cross-validation; Validierung mit Cox Proportional Hazards und Kaplan–Meier-Analyse in zwei Kohorten.
Ergebnisse
Herzinsuffizienz: Alter, Serumkreatinin und Blutdruck trennten Risikogruppen; Hochrisiko mit höherer Mortalität (HR 2,61; 95 %-KI 1,42–4,78; p = 0,0013). METABRIC: Alter bei Diagnose, HER2 und NPI zeigten starke Trennung; HR 2,73 (95 %-KI 2,34–3,19), medianes Überleben 104,7 vs. 252,3 Monate.
Schlussfolgerung der Autor:innen
Die Kombination aus cross-validierter ML-Feature-Selektion und Cox-basierter Überlebensanalyse ermöglicht eine stabile und klinisch interpretierbare Risikostratifizierung über verschiedene Krankheiten hinweg.
Kernaussagen
- Herzinsuffizienz: Hochrisiko mit HR 2,61 (95 %-KI 1,42–4,78)
- METABRIC: HR 2,73 (95 %-KI 2,34–3,19)
- C-index 0,665–0,794 über beide Kohorten
Deutsche Zusammenfassung: KI-gestützt aus dem Original-Abstract erstellt; alle Zahlen wurden automatisch mit dem Original abgeglichen. Maßgeblich ist der Original-Abstract.
Kurz gefasst
Fachartikel in der Zeitschrift Diagnostics, online erschienen am 06.03.2026. Verfasst von Fangya Tan und 6 weiteren Autor:innen. Beteiligte Einrichtung: Department of Analytics, Harrisburg University of Science and Technology, Harrisburg, PA 17101, USA. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 54 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Themen: Künstliche Intelligenz in der Pflege, Herzinsuffizienz
Abstract (Originalsprache)
Background
The clinical use of machine learning (ML) in survival analysis is often limited by the “black box” nature of complex algorithms, which makes their results difficult to interpret in practice. In this study, we propose a unified and clinically grounded framework that integrates ML-based feature selection with traditional survival analysis. This approach aims to bridge the gap between strong predictive performance and clear, clinically meaningful interpretation.
Methods
High-impact prognostic clinical features were identified using ML models GBM-Cox, RSF, and LASSO-Cox with 5-fold stratified cross-validation and subsequently validated using Cox Proportional Hazards and Kaplan–Meier analysis. The framework was evaluated across two distinct disease domains, Heart Failure and the METABRIC breast cancer cohort, to assess robustness and generalizability.
Results
In the Heart Failure dataset, age group, serum creatinine, and blood pressure stratified patients into distinct risk groups. The high-risk group had significantly higher mortality (HR: 2.61; 95% CI: 1.42–4.78; p = 0.0013). In the METABRIC cohort, age at diagnosis, HER2 status, and Nottingham Prognostic Index (NPI) showed strong survival separation (p < 0.001). The high-risk group had an HR of 2.73 (95% CI: 2.34–3.19) and the faced a significantly shorter median survival (104.7 vs. 252.3 months), representing a 12.3-year reduction in life expectancy compared to low-risk group. This prognostic separation emphasizes the predictive power of selected baseline variables. Performance remained stable across cohorts, with C-index values (0.665–0.794) consistent with standard clinical benchmarks.
Conclusions
Integrating cross-validated machine learning feature selection with Cox-based survival analysis enables stable and clinically interpretable risk stratification across diseases. By translating ML selected predictors into hazard ratios and absolute survival differences, this framework provides a reproducible and clinically grounded approach for survival risk assessment.
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Department of Analytics, Harrisburg University of Science and Technology, Harrisburg, PA 17101, USA.
- Department of Oncology, The Second Affiliated Hospital of Zunyi Medical University, Zunyi 563000, China.
- Department of Physics, The Graduate Center, City University of New York, New York, NY 10016, USA.
ORCID-Profile: Jian-Guo Zhou, Yang Zhou
Zitieren
APA
Tan, F., Zhou, J. G., Li, S., Long, B., Bellur, S., Zhou, Y., & Newman, M. (2026). A Unified Framework for Survival Prediction: Combining Machine Learning Feature Selection with Traditional Survival Analysis in Heart Failure and METABRIC Breast Cancer. Diagnostics, 16(5), 790. https://doi.org/10.3390/diagnostics16050790
Vancouver
Tan F, Zhou JG, Li S, Long B, Bellur S, Zhou Y, et al. A Unified Framework for Survival Prediction: Combining Machine Learning Feature Selection with Traditional Survival Analysis in Heart Failure and METABRIC Breast Cancer. Diagnostics. 2026;16(5):790. doi:10.3390/diagnostics16050790
BibTeX
@article{Tan2026,
author = {Tan, Fangya and Zhou, Jian-Guo and Li, Shuqiao and Long, Bowen and Bellur, Srikar and Zhou, Yang and Newman, Mark},
title = {A Unified Framework for Survival Prediction: Combining Machine Learning Feature Selection with Traditional Survival Analysis in Heart Failure and METABRIC Breast Cancer},
journal = {Diagnostics},
year = {2026},
volume = {16},
number = {5},
pages = {790},
doi = {10.3390/diagnostics16050790}
}
EUVIMED