Neural network and machine learning-based analysis of clinical associated features in breast cancer pathogenesis
Analyse klinisch assoziierter Merkmale bei der Pathogenese von Brustkrebs mit neuronalen Netzen und Machine Learning
Technology and Health Care · 04.11.2026
DOI: 10.1177/09287329261505770 ↗Auf einen Blick
In 460 Brustkrebspatient:innen wurde ein MLP mit anderen Modellen zur Analyse klinischer und Lebensstil-Faktoren geprüft. Das MLP erreichte eine Testgenauigkeit von 0,954 (95 %-KI 0,937–0,972).
Fragestellung
Lässt sich mit einem neuronalen Netz klinisch assoziierte Merkmale bei Brustkrebs validieren und interpretierbar analysieren?
Methode
MLP mit Logistischer Regression, Random Forest und XGBoost; Daten von 460 Brustkrebspatient:innen mit 17 Merkmalen; SMOTE, permutation feature importance, SHAP, ANOVA, multivariable logistische Regression; Stratified 5-Fold Cross-Validation.
Ergebnisse
MLP: Testgenauigkeit 0,954 (95 %-KI 0,937–0,972). Logistische Regression: höchste mittlere AUC 0,861. Random Forest: Genauigkeit 0,972. Alkoholgebrauch war unabhängig mit hoher Schwere assoziiert: adj. OR 10,751 (95 %-KI 2,545–45,422; p = 0,001).
Schlussfolgerung der Autor:innen
Das Framework identifiziert und validiert laut Autor:innen kritische Brustkrebs-assoziierte Merkmale und bietet einen interpretierbaren computationalen Ansatz zur Risikostratifizierung.
Kernaussagen
- MLP-Testgenauigkeit: 0,954 (95 %-KI 0,937–0,972)
- Random Forest mit Genauigkeit 0,972
- Alkoholgebrauch: adj. OR 10,751 (95 %-KI 2,545–45,422)
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 Technology and Health Care, online erschienen am 04.11.2026. Verfasst von Lijing Yang und 1 weiteren Autor:innen. Beteiligte Einrichtung: Wuhan Children's Hospital (Wuhan Maternal and Child Healthcare Hospital), Tongji Medical College, Huazhong University of Science & Technology, Wuhan,…. Die Arbeit nennt 30 Literaturangaben.
Abstract (Originalsprache)
Background
Breast cancer remains one of the most prevalent malignancies worldwide, with complex multifactorial etiology involving genetic, environmental, and lifestyle determinants. Early identification of critical associated features within confirmed patient cohorts through computational methods could significantly enhance clinical management and clinical decision-making.
Objective
This study aims to develop and validate a neural network framework for analyzing breast cancer associated clinical features, with particular focus on interpretable feature importance analysis and statistical validation of identified associations.
Methods
We employed a multi-layer perceptron (MLP) neural network alongside baseline models (Logistic Regression, Random Forest, XGBoost) trained on comprehensive clinical and lifestyle data from 460 breast cancer patients. To optimize clinical relevance and address class imbalance, target variables were binarized and SMOTE was applied. The dataset included 17 features encompassing demographic, genetic, behavioral, and environmental factors. Model interpretability was enhanced through permutation feature importance and SHAP analysis, while statistical significance was validated via ANOVA testing and multivariable logistic regression. Performance was evaluated using standard classification metrics and Stratified 5-Fold Cross-Validation techniques.
Results
Under stratified fivefold cross-validation, the MLP achieved overall test accuracy of 0.954 (95% CI: 0.937–0.972). Logistic regression had the highest mean AUC (0.861), while Random Forest achieved accuracy of 0.972. In multivariable logistic regression, alcohol use showed the strongest independent association with high-risk severity (adjusted OR = 10.751, 95% CI: 2.545–45.422; p = 0.001).
Conclusion
The presented framework successfully identifies and validates critical breast cancer associated features, providing an interpretable computational approach for risk stratification. The integration of neural networks with statistical validation and SHAP analysis offers a robust methodology for clinical risk assessment and personalized patient management strategies.
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Wuhan Children's Hospital (Wuhan Maternal and Child Healthcare Hospital), Tongji Medical College, Huazhong University of Science & Technology, Wuhan, China.
Zitieren
APA
Yang, L., & Wang, W. (2026). Neural network and machine learning-based analysis of clinical associated features in breast cancer pathogenesis. Technology and Health Care. https://doi.org/10.1177/09287329261505770
Vancouver
Yang L, Wang W. Neural network and machine learning-based analysis of clinical associated features in breast cancer pathogenesis. Technology and Health Care. 2026. doi:10.1177/09287329261505770
BibTeX
@article{Yang2026,
author = {Yang, Lijing and Wang, Wanju},
title = {Neural network and machine learning-based analysis of clinical associated features in breast cancer pathogenesis},
journal = {Technology and Health Care},
year = {2026},
doi = {10.1177/09287329261505770}
}
EUVIMED