Hybrid Computer Vision Model to Predict Lung Cancer in Diverse Populations
- 1 January 2026
- journal article
- research article
- Published by American Society of Clinical Oncology (ASCO) in JCO Clinical Cancer Informatics
- Vol. 10 (1) , e2500041
Abstract
PURPOSE: Disparities in lung cancer incidence exist in Black populations, and screening criteria underserve Black populations due to disparately elevated risk in the screening-eligible population. Prediction models that integrate clinical and imaging-based features to individualize lung cancer risk are a potential means to mitigate these disparities. METHODS: This multicenter (National Lung Screening Trial [NLST]) and catchment population–based (University of Illinois Health [UIH], urban and suburban Cook County) cross-sectional study used participants at risk of lung cancer with available lung computed tomography (CT) imaging and follow-up between the years 2015 and 2024. In all, 53,452 in NLST and 11,654 in UIH were included on the basis of age and tobacco use–based risk factors for lung cancer. Cohorts were used for training and testing of deep and machine learning models using clinical features alone or combined with CT image features (hybrid computer vision). RESULTS: An optimized seven-feature clinical model achieved receiver operating characteristic (ROC)-AUC values ranging from 0.64 to 0.67 in NLST and 0.60 to 0.65 in UIH cohorts across multiple years. Incorporation of imaging features to form a hybrid computer vision model significantly improved ROC-AUC values to 0.78-0.91 in NLST but deteriorated in UIH with ROC-AUC values of 0.68-0.80, attributable to Black participants where ROC-AUC values ranged from 0.63 to 0.72 across multiple years. Retraining the hybrid computer vision model by incorporating Black and other participants from the UIH cohort improved performance with ROC-AUC values of 0.70-0.87 in a held-out UIH test set. CONCLUSION: Hybrid computer vision predicted risk with improved accuracy compared with clinical risk models alone. However, potential biases in image training data reduced model generalizability in Black participants. Performance was improved upon retraining with a subset of the UIH cohort, suggesting that inclusive training and validation data sets can minimize racial disparities. Future studies incorporating vision models trained on representative data sets may demonstrate improved health equity upon clinical use.Keywords
This publication has 33 references indexed in Scilit:
- Identifying and Addressing Bias in Artificial IntelligenceJAMA Network Open, 2024
- Lung Cancer Screening in the US, 2022JAMA Internal Medicine, 2024
- Risk Model–Based Lung Cancer Screening and Racial and Ethnic Disparities in the USJAMA Oncology, 2023
- Methods for Using Race and Ethnicity in Prediction Models for Lung Cancer Screening EligibilityJAMA Network Open, 2023
- “Shortcuts” Causing Bias in Radiology Artificial Intelligence: Causes, Evaluation, and MitigationJournal of the American College of Radiology, 2023
- Potential Disparities by Sex and Race or Ethnicity in Lung Cancer Screening Eligibility RatesChest, 2021
- Risk Prediction Model Versus United States Preventive Services Task Force Lung Cancer Screening Eligibility Criteria: Reducing Race DisparitiesJournal of Thoracic Oncology, 2020
- Racial and Ethnic Disparities in Early-Stage Lung Cancer SurvivalChest, 2017
- Development and Validation of Risk Models to Select Ever-Smokers for CT Lung Cancer ScreeningJAMA, 2016
- Misuse of DeLong test to compare AUCs for nested modelsStatistics in Medicine, 2012