Novel models based on machine learning to predict the prognosis of metaplastic breast cancer
Contact us if these videos are not relevant.
Contact us if these videos are not relevant.
View abstract on PubMed
Summary
This summary is machine-generated.A new CatBoost machine learning model accurately predicts survival for metaplastic breast cancer (MBC) patients. Radiotherapy is more beneficial for specific patient groups undergoing breast-conserving surgery or mastectomy.
Area Of Science
- Oncology
- Machine Learning
- Biostatistics
Background
- Metaplastic breast cancer (MBC) is a rare, aggressive subtype with limited predictive models.
- Accurate prognostication is crucial for effective clinical management of MBC.
Purpose Of The Study
- To develop and validate a machine learning model for predicting survival in MBC patients.
- To evaluate the efficacy of radiotherapy in different MBC patient subgroups.
Main Methods
- Utilized SEER database (2010-2018) and a hospital cohort for model development and validation.
- Developed a novel CatBoost machine learning model incorporating prognostic factors.
- Compared radiotherapy benefits across patient groups based on surgical approach and disease stage.
Main Results
- The CatBoost model demonstrated high accuracy in predicting MBC patient survival (e.g., 1-year AUC=0.833, 5-year AUC=0.810).
- The model showed strong performance on an independent dataset (e.g., 1-year AUC=0.937, 5-year AUC=0.890).
- Radiotherapy showed significant survival benefits for patients with M0 stage undergoing breast-conserving surgery and for T3-4/N2-3M0 stage patients undergoing mastectomy.
Conclusions
- A validated CatBoost model can effectively predict survival outcomes in MBC.
- Radiotherapy is recommended for specific MBC patient groups: those with M0 stage post-breast-conserving surgery and T3-4/N2-3M0 stage post-mastectomy.
Contact us if these videos are not relevant.
Contact us if these videos are not relevant.