Related Experiment Video
Updated: Aug 6, 2026
07:15
Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model
Published on: August 16, 2020
Development and validation of an interpretable machine learning-based predictive model for breast cancer bone
Caiyun Fan1, Ming Tian1, Zhendong Ding2,3
1Department of Anesthesiology, The First People's Hospital of Kashi, Kashi, China.
Frontiers in Oncology
|July 23, 2026
Summary
This study developed a machine learning model to predict bone metastasis in breast cancer patients, identifying key risk factors like age and tumor stage. The model aids in personalized treatment and early detection of metastasis.
Area of Science:
- Oncology
- Biostatistics
- Machine Learning
Background:
- Breast cancer is a prevalent malignancy with bone metastasis significantly impacting patient prognosis.
- Bone metastasis is the most common form of distant spread in breast cancer patients.
- Accurate risk assessment is crucial for personalized intervention and treatment optimization.
Purpose of the Study:
- To develop a machine learning-based predictive model for bone metastasis risk in breast cancer.
- To enable personalized risk stratification and early clinical intervention.
- To optimize treatment strategies for breast cancer patients at risk of bone metastasis.
Main Methods:
- Utilized the Surveillance, Epidemiology, and End Results (SEER) database for model development.
- Employed logistic regression for variable screening and eight machine learning algorithms for prediction.
- Validated models using internal and external cohorts, assessing performance with AUC, AUPRC, and SHAP analysis.
Main Results:
- The LGB model achieved high predictive performance (AUC 0.98 training, 0.91 external validation).
- Key risk factors identified include age >50, higher tumor grade, advanced T/N stage, and HR-/HER2- subtype.
- Surgery was identified as a primary protective factor, while advanced N stage increased risk.
Conclusions:
- Developed an interpretable LGB model with a web-based calculator for clinical application.
- The model facilitates personalized risk stratification for breast cancer patients.
- Aids in early detection of bone metastasis and optimization of treatment strategies.