EUVIMEDEUVIMED entdecken
Multizentrische Studie

Hybrid Computer Vision Model to Predict Lung Cancer in Diverse Populations

Abdul J Zakkar, Nazia Perwaiz, Vikram Harikrishnan, Weiheng Zhong iD, Vijeth Narra, Farah Yousef, Daniel Kim, Mason Burrage-Burton, Abdul Afeez Lawal iD, Vijayakrishna K Gadi iD, Mark C Korpics iD, Sage J Kim iD und 8 weitere

JCO Clinical Cancer Informatics · 2026-01 · 10 (1)

DOI: 10.1200/cci-25-00041 ↗

Kurz gefasst

Multizentrische Studie in der Zeitschrift JCO Clinical Cancer Informatics, 2026-01. Verfasst von Abdul J Zakkar und 19 weiteren Autor:innen. Beteiligte Einrichtung: Department of Medicine, College of Medicine, University of Illinois Chicago, Chicago, IL. Hauptthemen laut MEDLINE: Lung Neoplasms. Die Arbeit nennt 38 Literaturangaben.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

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.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Lung NeoplasmsHumansTomography, X-Ray ComputedFemaleEarly Detection of CancerCross-Sectional StudiesMachine LearningPredictive Learning ModelsRisk FactorsROC Curve

Autor:innen und Institutionen

  • Department of Medicine, College of Medicine, University of Illinois Chicago, Chicago, IL.
  • Department of Biochemistry and Molecular Genetics, College of Medicine, University of Illinois Chicago, Chicago, IL.
  • School of Electrical Engineering and Computer Science, National University of Science and Technology, Islamabad, Pakistan.
  • Department of Civil and Environmental Engineering, Grainger College of Engineering, University of Illinois Urbana Champaign, Urbana Champaign, IL.
  • Department of Radiation Oncology, College of Medicine, University of Illinois Chicago, Chicago, IL.
  • School of Public Health, University of Illinois Chicago, Chicago, IL.

ORCID-Profile: Weiheng Zhong, Abdul Afeez Lawal, Vijayakrishna K Gadi, Mark C Korpics, Sage J Kim, Zhengjia Chen, Aly A Khan, Yamilé Molina, Yang Dai, Hadi Meidani, Ryan H Nguyen, Ameen A Salahudeen

Zitieren

APA

Zakkar, A. J., Perwaiz, N., Harikrishnan, V., Zhong, W., Narra, V., Yousef, F., Kim, D., Burrage-Burton, M., Lawal, A. A., Gadi, V. K., Korpics, M. C., Kim, S. J., Chen, Z., Khan, A. A., Molina, Y., Dai, Y., Marai, G. E., Meidani, H., Nguyen, R. H., & Salahudeen, A. A. (2026). Hybrid Computer Vision Model to Predict Lung Cancer in Diverse Populations. JCO Clinical Cancer Informatics, 10(1). https://doi.org/10.1200/cci-25-00041

Vancouver

Zakkar AJ, Perwaiz N, Harikrishnan V, Zhong W, Narra V, Yousef F, et al. Hybrid Computer Vision Model to Predict Lung Cancer in Diverse Populations. JCO Clinical Cancer Informatics. 2026;10(1). doi:10.1200/cci-25-00041

BibTeX
@article{Zakkar2026,
  author = {Zakkar, Abdul J and Perwaiz, Nazia and Harikrishnan, Vikram and Zhong, Weiheng and Narra, Vijeth and Yousef, Farah and Kim, Daniel and Burrage-Burton, Mason and Lawal, Abdul Afeez and Gadi, Vijayakrishna K and Korpics, Mark C and Kim, Sage J and Chen, Zhengjia and Khan, Aly A and Molina, Yamilé and Dai, Yang and Marai, G Elisabeta and Meidani, Hadi and Nguyen, Ryan H and Salahudeen, Ameen A},
  title = {Hybrid Computer Vision Model to Predict Lung Cancer in Diverse Populations},
  journal = {JCO Clinical Cancer Informatics},
  year = {2026},
  volume = {10},
  number = {1},
  doi = {10.1200/cci-25-00041}
}
Kostenlos starten →