Related Experiment Video
Updated: Jun 29, 2026
Author Spotlight: Advancing 3D Modeling for Enhanced Diagnosis and Treatment of Pulmonary Nodules in Early-Stage Lung Cancer
Published on: October 13, 2023
Prediction of the Preclinical Stage of Coal Workers' Pneumoconiosis on Nonimaging Data Integrating Prior Knowledge
Fengtao Cui1, Hui Xu, Yankun Ma
1From the Occupational Health Surveillance and Management Center, Occupational Disease Prevention and Control Institute, Huaibei Mining Co., Ltd., Huaibei, Anhui, China (F.C.); Occupational Health Assessment Department, Jinneng Holding Coal Industry Group Co., Ltd. Occupational Disease Prevention and Treatment Hospital, DaTong, Shanxi, China (H.X.); Key Laboratory of Environmental Stress and Chronic Disease Control and Prevention, Ministry of Education (China Medical University), Shenyang, Liaoning, China (Y.M., B.L.); Key Laboratory of Arsenic-related Biological Effects and Prevention and Treatment in Liaoning Province (China Medical University), Shenyang, Liaoning, China (Y.M., B.L.); Environment and Non-Communicable Disease Research Center, School of Public Health, China Medical University, Shenyang, Liaoning, China (Y.M., B.L.); CT Department, Shenyang Ninth People's Hospital, Shenyang, Liaoning, China (K.H.); School of Public Health, North China University of Science and Technology, Tangshan, Hebei, China (F.S.); and Department of Biomedical Engineering, School of Intelligent Medicine, China Medical University, Shenyang, Liaoning, China (Y.W.).
Objective:
This study aims to establish machine learning models using nonimaging data from health examinations of coal workers, which can screen the preclinical stage of coal workers' pneumoconiosis (CWP).
Methods:
Nonimaging data from two centers, totaling 34,362 coal miners, were collected. From 84 initial variables, 19 were preliminarily screened, and LASSO selected eight key features. Six machine learning models were trained to predict the preclinical stage of CWP, evaluated using ROC curve.
Results:
In the internal test set, GB achieved the best discrimination (AUC 88.19%), while DT yielded the highest accuracy (81.09%) and specificity (80.90%). In the external validation set, GB remained the top model by AUC (83.94%) and showed high sensitivity (87.67%).
Conclusions:
Age, FEV1, FEV1%, drinking status, smoking status, FVC, occupational category, and cumulative years of service are significant features for predicting the preclinical stage of CWP.
More Related Videos
03:38Unilateral Lung Volume Analysis Using Micro-CT for Enhanced Assessment of Pulmonary Fibrosis in Preclinical Models
Published on: June 20, 2025
06:22Machine Learning-Based Cough Tone Classification: Diagnostic Exploration of Chronic Obstructive Pulmonary Disease and Respiratory Tract Infections
Published on: September 19, 2025
