SARS-Cov-2 pneumonia phenotyping on imaging exams of patients submitted to minimally invasive autopsy.

Background: Correlation between pathology and imaging of the new SARS-Cov-2 disease (COVID-19) is scarce. This study aimed to characterize SARS-Cov-2 pneumonia on imaging of patients submitted to minimally invasive autopsy (MIA). Methods: This unicentric retrospective observational study included 46 consecutive patients with confirmed COVID-19 who underwent MIA. All clinical chest images were reviewed and classified for the presence and grade of viral pneumonia, as well as disease evolution. On CT, phenotypes were described as consistent with mild, moderate, or severe viral pneumonia, with or without radiological signs of organizing pneumonia (OP). In severe pneumonia, CT could also be classified as diffuse progressive OP or radiological diffuse alveolar damage (DAD). Specific features on CT were noted, including fibroproliferative signs that could indicate potential or initial fibrosis. Results: MIA showed a heterogeneous panel of alterations, with a high prevalence of OP and acute fibrinous and organizing pneumonia (AFOP). Also, signs of interstitial fibrosis corresponded to the most prevalent pathological feature. Initial chest radiography (CXR) findings were mainly consistent with moderate or severe viral pneumonia. Most patients showed stability or improvement (reduction of opacities) on imaging. CTs were performed on 15 patients. Consolidations were found in most patients, frequently showing features consistent with an OP phenotype. Fibroproliferative changes were also prevalent on CT. Conclusions: In this study, SARS-Cov-2 pneumonia showed heterogeneous radiological and pathological patterns. Signs of organization and potential or initial fibrosis were prevalent on both imaging and pathology. Imaging phenotyping may help to predict post-infection fibrosing interstitial pneumonitis in COVID-19.

COVID-19 pneumonia accurately detected on chest radiographs with artificial intelligence.

PURPOSE: To investigate the diagnostic performance of an Artificial Intelligence (AI) system for detection of COVID-19 in chest radiographs (CXR), and compare results to those of physicians working alone, or with AI support. MATERIALS AND METHODS: An AI system was fine-tuned to discriminate confirmed COVID-19 pneumonia, from other viral and bacterial pneumonia and non-pneumonia patients and used to review 302 CXR images from adult patients retrospectively sourced from nine different databases. Fifty-four physicians blind to diagnosis, were invited to interpret images under identical conditions in a test set, and randomly assigned either to receive or not receive support from the AI system. Comparisons were then made between diagnostic performance of physicians working with and without AI support. AI system performance was evaluated using the area under the receiver operating characteristic (AUROC), and sensitivity and specificity of physician performance compared to that of the AI system. RESULTS: Discrimination by the AI system of COVID-19 pneumonia showed an AUROC curve of 0.96 in the validation and 0.83 in the external test set, respectively. The AI system outperformed physicians in the AUROC overall (70% increase in sensitivity and 1% increase in specificity, p < 0.0001). When working with AI support, physicians increased their diagnostic sensitivity from 47% to 61% (p < 0.001), although specificity decreased from 79% to 75% (p = 0.007). CONCLUSIONS: Our results suggest interpreting chest radiographs (CXR) supported by AI, increases physician diagnostic sensitivity for COVID-19 detection. This approach involving a human-machine partnership may help expedite triaging efforts and improve resource allocation in the current crisis.