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.
Pyrros, Ayis; Borstelmann, Stephen M.; Mantravadi, Ramana; Zaiman, Zachary; Thomas, Kaesha; Price, Brandon; Greenstein, Eugene; Siddiqui, Nasir; Willis, Melinda; Shulhan, Ihar; Hines-Shah, John; Horowitz, Jeanne M.; Nikolaidis, Paul; Lungren, Matthew P.; Rodríguez-Fernández, Jorge Mario; Gichoya, Judy Wawira; Koyejo, Sanmi; Flanders, Adam E.; Khandwala, Nishith; Gupta, Amit; Garrett, John W.; Cohen, Joseph Paul; Layden, Brian T.; Pickhardt, Perry J.; and Galanter, William, "Opportunistic Detection of Type 2 Diabetes Using Deep Learning From Frontal Chest Radiographs" (2023). Department of Radiology Faculty Papers. Paper 151.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Peer Review.pdf (3542 kB)
Peer Review File
41467_2023_39631_MOESM3_ESM.docx (14 kB)
Description of Additional Supplementary Information File
41467_2023_39631_MOESM4_ESM.mp4 (32666 kB)
Supplementary Movie 1
Reporting Summary.pdf (1013 kB)