Machine learning-based prediction of distant metastasis risk in invasive ductal carcinoma of the breast

Jingru Dong, Ruijiao Lei, Feiyang Ma, Lu Yu, Lanlan Wang, Shangzhi Xu, Yunhua Hu, Jialin Sun, Wenwen Zhang, Haixia Wang and Li Zhang

PLOS ONE, 2025, vol. 20, issue 2, 1-21

Abstract: More than 90% of deaths due to breast cancer (BC) are due to metastasis-related complications, with invasive ductal carcinoma (IDC) of the breast being the most common pathologic type of breast cancer and highly susceptible to metastasis to distant organs. BC patients who develop cancer metastases are more likely to have a poor prognosis and poor quality of life, so it is extremely important to recognize and diagnose whether distant metastases have occurred in IDC as early as possible. In this study, we develop a non-invasive breast cancer classification system for detecting cancer metastasis. We used Anaconda-Jupyter notebooks to develop various Python programming modules for text mining, data processing, and machine learning (ML) methods. A risk prediction model was constructed based on four algorithms: Random Forest, XGBoost, Logistic Regression, and SVM. Additionally, we developed a hybrid model based on a voting mechanism using these four algorithms as the base models. The models were compared and evaluated by the following metrics: accuracy, precision, recall, F1-score, and area under the ROC curve (AUC) values. The experimental results show that the hybrid model based on the voting mechanism exhibits the best prediction performance (accuracy: 0.867, precision: 0.929, recall: 0.805, F1-score: 0.856, AUC: 0.94). This stable risk prediction model provides a valuable reference support for doctors in assessing and diagnosing the risk of IDC hematogenous metastasis. It also improves the work efficiency of doctors and strives to provide patients with increased chances of survival.

Date: 2025
References: Add references at CitEc
Citations:

Downloads: (external link)
https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0310410 (text/html)
https://journals.plos.org/plosone/article/file?id= ... 10410&type=printable (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:plo:pone00:0310410

DOI: 10.1371/journal.pone.0310410

Access Statistics for this article

More articles in PLOS ONE from Public Library of Science
Bibliographic data for series maintained by plosone ().