Artificial intelligence helpful in stratifying COVID-19 risk based on chest x-rays: Study
USA: A modified commercially available deep learning algorithm (M-qXR) could be utilized in low-resource settings to risk-stratify patients with suspected COVID-19 infections by detecting abnormalities on chest X-rays, shows a recent study. The study was published in the journal Intelligence-Based Medicine on January 13, 2022. According to the study, the artificial intelligence...
USA: A modified commercially available deep learning algorithm (M-qXR) could be utilized in low-resource settings to risk-stratify patients with suspected COVID-19 infections by detecting abnormalities on chest X-rays, shows a recent study. The study was published in the journal Intelligence-Based Medicine on January 13, 2022.
According to the study, the artificial intelligence (AI) algorithm had comparable accuracy to ground truth for detecting radiographic abnormalities on CXR suggestive of COVID-19 and thus could serve as a radiology decision tool to guide management of patients who are deemed at risk for COVID-19.
Previous studies have shown that in resource-limited settings, deep learning-based radiological image analysis could facilitate the use of chest x-rays as a triaging tool for COVID-19 diagnosis. Diego A. Hipolito Canario, UNC School of Medicine, the University of North Carolina at Chapel Hill, Chapel Hill, NC, United States, and colleagues thus aimed to determine whether a modified commercially available deep learning algorithm (M-qXR) could risk-stratify patients with suspected COVID-19 infections.
For this purpose, the researchers designed a dual-track clinical validation study to assess the clinical accuracy of M-qXR. All Chest-X-rays (CXRs) performed during the study period were evaluated for abnormal findings and assigned a COVID-19 risk score. Four independent radiologists served as radiological ground truth.
The researchers then compared M-qXR algorithm output against radiological ground truth and calculated summary statistics for prediction accuracy. In a co-occurrence matrix, patients who underwent both PCR testing and CXR for suspected COVID-19 infection were included to assess the sensitivity and specificity of the M-qXR algorithm.
The study revealed the following findings:
- 625 CXRs were included in the clinical validation study. 98% of total interpretations made by M-qXR agreed with ground truth.
- 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.
- 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.
To conclude, M-qXR was shown to have comparable accuracy to radiological ground truth in detecting radiographic abnormalities on CXR suggestive of COVID-19.
Reference:
The study titled, "Using artificial intelligence to risk stratify COVID-19 patients based on chest X-ray findings," was published in the journal Intelligence-Based Medicine.
