Nature Communications (Jul 2023)

Opportunistic detection of type 2 diabetes using deep learning from frontal chest radiographs

  • Ayis Pyrros,
  • Stephen M. Borstelmann,
  • Ramana Mantravadi,
  • Zachary Zaiman,
  • Kaesha Thomas,
  • Brandon Price,
  • Eugene Greenstein,
  • Nasir Siddiqui,
  • Melinda Willis,
  • Ihar Shulhan,
  • John Hines-Shah,
  • Jeanne M. Horowitz,
  • Paul Nikolaidis,
  • Matthew P. Lungren,
  • Jorge Mario Rodríguez-Fernández,
  • Judy Wawira Gichoya,
  • Sanmi Koyejo,
  • Adam E Flanders,
  • Nishith Khandwala,
  • Amit Gupta,
  • John W. Garrett,
  • Joseph Paul Cohen,
  • Brian T. Layden,
  • Perry J. Pickhardt,
  • William Galanter

DOI
https://doi.org/10.1038/s41467-023-39631-x
Journal volume & issue
Vol. 14, no. 1
pp. 1 – 12

Abstract

Read online

Abstract Deep learning (DL) models can harness electronic health records (EHRs) to predict diseases and extract radiologic findings for diagnosis. With ambulatory chest radiographs (CXRs) frequently ordered, we investigated detecting type 2 diabetes (T2D) by combining radiographic and EHR data using a DL model. Our model, developed from 271,065 CXRs and 160,244 patients, was tested on a prospective dataset of 9,943 CXRs. Here we show the model effectively detected T2D with a ROC AUC of 0.84 and a 16% prevalence. The algorithm flagged 1,381 cases (14%) as suspicious for T2D. External validation at a distinct institution yielded a ROC AUC of 0.77, with 5% of patients subsequently diagnosed with T2D. Explainable AI techniques revealed correlations between specific adiposity measures and high predictivity, suggesting CXRs’ potential for enhanced T2D screening.