Construction of a Lung Cancer Screening Risk Prediction Model Based on Machine Learning Algorithms
Tiantian Zhang1,2, Yexin Chen1, Pei Wang3,4
1College of Pharmacy/Southern Institute of Pharmacoeconomics and Health Technology Assessment, Jinan University, Guangzhou, China.
Machine learning models, like XGBoost, can improve lung cancer risk prediction for screening in China. The XGBoost model showed better accuracy than logistic regression, highlighting childhood cooking fuel exposure as a risk factor.
Area of Science:
- Oncology
- Data Science
- Public Health
Background:
- Lung cancer screening accuracy can be improved by risk prediction models.
- Research on these models, especially machine learning, is limited in China.
Purpose of the Study:
- To develop and compare machine learning-based lung cancer risk prediction models for screening in China.
- To identify key risk factors for lung cancer in the Chinese population.
Main Methods:
- A prospective cohort of 11,708 participants was used.
- LASSO regression screened variables.
- Logistic regression and XGBoost models were built and validated.
Main Results:
- The XGBoost model (AUC: 0.658) demonstrated slightly better discriminative ability than logistic regression (AUC: 0.647).
- Childhood exposure to cooking fuels was identified as a significant, understudied risk factor for lung cancer.
Conclusions:
- The XGBoost model offers superior prediction accuracy and robustness for lung cancer risk assessment during screening.
- This study provides a valuable tool for identifying high-risk individuals in China.
More Related Videos
07:15Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model
Published on: August 16, 2020
04:09Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma
Published on: October 10, 2018
