Using artificial intelligence to risk stratify COVID-19 patients based on chest X-ray findings

Diego A Hipolito Canario; Eric Fromke; Matthew A Patetta; Mohamed T Eltilib; Juan P Reyes-Gonzalez; Georgina Cornelio Rodriguez; Valeria A Fusco Cornejo; Seymour Dunckner; Jessica K Stewart

doi:10.1016/j.ibmed.2022.100049

Using artificial intelligence to risk stratify COVID-19 patients based on chest X-ray findings

Intell Based Med. 2022:6:100049. doi: 10.1016/j.ibmed.2022.100049. Epub 2022 Jan 13.

Authors

Diego A Hipolito Canario¹, Eric Fromke¹, Matthew A Patetta¹, Mohamed T Eltilib¹, Juan P Reyes-Gonzalez², Georgina Cornelio Rodriguez², Valeria A Fusco Cornejo³, Seymour Dunckner³, Jessica K Stewart⁴

Affiliations

¹ UNC School of Medicine, University of North Carolina at Chapel Hill, Bondurant Hall, CB 9535, Chapel Hill, NC, 27599-3280, United States.
² Department of Radiology, Angeles del Pedregal Hospital, Camino de Sta, Teresa 1055-S, Héroes de Padierna, La Magdalena Contreras, 10700, Ciudad de México, Mexico.
³ Mindscale, 800 W El Camino Real Suite 180 Mountain View, CA, 94040, United States.
⁴ Division of Interventional Radiology, Department of Radiology, David Geffen School of Medicine, University of California at Los Angeles. 757 Westwood Plaza, Suite 2125, Los Angeles, CA, 90095, United States.

Abstract

Background: Deep learning-based radiological image analysis could facilitate use of chest x-rays as a triaging tool for COVID-19 diagnosis in resource-limited settings. This study sought to determine whether a modified commercially available deep learning algorithm (M-qXR) could risk stratify patients with suspected COVID-19 infections.

Methods: A dual track clinical validation study was designed to assess the clinical accuracy of M-qXR. The algorithm evaluated all Chest-X-rays (CXRs) performed during the study period for abnormal findings and assigned a COVID-19 risk score. Four independent radiologists served as radiological ground truth. The M-qXR algorithm output was compared against radiological ground truth and summary statistics for prediction accuracy were calculated. In addition, patients who underwent both PCR testing and CXR for suspected COVID-19 infection were included in a co-occurrence matrix to assess the sensitivity and specificity of the M-qXR algorithm.

Results: 625 CXRs were included in the clinical validation study. 98% of total interpretations made by M-qXR agreed with ground truth (p = 0.25). M-qXR correctly identified the presence or absence of pulmonary opacities in 94% of CXR interpretations. M-qXR's sensitivity, specificity, PPV, and NPV for detecting pulmonary opacities were 94%, 95%, 99%, and 88% respectively. M-qXR correctly identified the presence or absence of pulmonary consolidation in 88% of CXR interpretations (p = 0.48). M-qXR's sensitivity, specificity, PPV, and NPV for detecting pulmonary consolidation were 91%, 84%, 89%, and 86% respectively. Furthermore, 113 PCR-confirmed COVID-19 cases were used to create a co-occurrence matrix between M-qXR's COVID-19 risk score and COVID-19 PCR test results. The PPV and NPV of a medium to high COVID-19 risk score assigned by M-qXR yielding a positive COVID-19 PCR test result was estimated to be 89.7% and 80.4% respectively.

Conclusion: M-qXR was found to have comparable accuracy to radiological ground truth in detecting radiographic abnormalities on CXR suggestive of COVID-19.

Keywords: AI, artificial intelligence; Artificial intelligence; COVID-19; COVID-19, coronavirus disease of 2019; CXR, chest x-rays; Chest X-ray; Deep learning algorithm; M-qXR, modified qXR deep learning algorithm; Patient risk stratification.