Development and external validation of a machine learning model for prediction of survival in extremity leiomyosarcoma
These videos have been matched automatically. Contact us if they are not relevant.
These videos have been matched automatically. Contact us if they are not relevant.
View abstract on PubMed
Summary
This summary is machine-generated.Machine learning models accurately predict survival for extremity leiomyosarcoma (LMS), a rare cancer. The developed Random Forest model offers a valuable tool for LMS prognostication and patient care.
Area Of Science
- Oncology
- Machine Learning
- Biostatistics
Background
- Machine learning (ML) models have shown promise in predicting cancer survival across various sarcoma subtypes.
- Extremity leiomyosarcoma (LMS) survival prediction has not been extensively studied using ML approaches.
- ML offers a powerful avenue for improved prognostication in extremity LMS.
Purpose Of The Study
- To develop and validate machine learning models for predicting survival in extremity leiomyosarcoma (LMS).
- To investigate the potential of ML in enhancing the prognostication of extremity LMS.
- To make the best-performing ML model publicly accessible for broader use.
Main Methods
- Utilized the Surveillance, Epidemiology, and End Results (SEER) database for 634 extremity LMS cases.
- Developed ML models to predict 1-, 3-, and 5-year survival based on patient, tumor, and treatment characteristics.
- Externally validated the top-performing ML model using an institutional cohort of 46 extremity LMS patients.
Main Results
- ML models demonstrated optimal performance at 1-year and reduced performance at 5-year survival prediction.
- Internal validation within the SEER cohort yielded c-statistics of 0.75-0.76 at 5 years.
- The Random Forest (RF) model achieved c-statistics of 0.90 (1-year) and 0.87 (5-year) on external validation, showing good calibration.
Conclusions
- Machine learning models exhibit excellent performance in predicting survival for extremity LMS.
- The developed RF model is a valuable tool for LMS prognostication.
- Further validation with larger institutional cohorts may enhance the ML model's utility for LMS prognostication.
These videos have been matched automatically. Contact us if they are not relevant.
These videos have been matched automatically. Contact us if they are not relevant.
