Deep Neural Network Augments Performance of Junior Residents in Diagnosing COVID-19 Pneumonia on Chest Radiographs.

Chest X-rays (CXRs) are essential in the preliminary radiographic assessment of patients affected by COVID-19. Junior residents, as the first point-of-contact in the diagnostic process, are expected to interpret these CXRs accurately. We aimed to assess the effectiveness of a deep neural network in distinguishing COVID-19 from other types of pneumonia, and to determine its potential contribution to improving the diagnostic precision of less experienced residents. A total of 5051 CXRs were utilized to develop and assess an artificial intelligence (AI) model capable of performing three-class classification, namely non-pneumonia, non-COVID-19 pneumonia, and COVID-19 pneumonia. Additionally, an external dataset comprising 500 distinct CXRs was examined by three junior residents with differing levels of training. The CXRs were evaluated both with and without AI assistance. The AI model demonstrated impressive performance, with an Area under the ROC Curve (AUC) of 0.9518 on the internal test set and 0.8594 on the external test set, which improves the AUC score of the current state-of-the-art algorithms by 1.25% and 4.26%, respectively. When assisted by the AI model, the performance of the junior residents improved in a manner that was inversely proportional to their level of training. Among the three junior residents, two showed significant improvement with the assistance of AI. This research highlights the novel development of an AI model for three-class CXR classification and its potential to augment junior residents' diagnostic accuracy, with validation on external data to demonstrate real-world applicability. In practical use, the AI model effectively supported junior residents in interpreting CXRs, boosting their confidence in diagnosis. While the AI model improved junior residents' performance, a decline in performance was observed on the external test compared to the internal test set. This suggests a domain shift between the patient dataset and the external dataset, highlighting the need for future research on test-time training domain adaptation to address this issue.

Contrastive domain adaptation with consistency match for automated pneumonia diagnosis.

Pneumonia can be difficult to diagnose since its symptoms are too variable, and the radiographic signs are often very similar to those seen in other illnesses such as a cold or influenza. Deep neural networks have shown promising performance in automated pneumonia diagnosis using chest X-ray radiography, allowing mass screening and early intervention to reduce the severe cases and death toll. However, they usually require many well-labelled chest X-ray images for training to achieve high diagnostic accuracy. To reduce the need for training data and annotation resources, we propose a novel method called Contrastive Domain Adaptation with Consistency Match (CDACM). It transfers the knowledge from different but relevant datasets to the unlabelled small-size target dataset and improves the semantic quality of the learnt representations. Specifically, we design a conditional domain adversarial network to exploit discriminative information conveyed in the predictions to mitigate the domain gap between the source and target datasets. Furthermore, due to the small scale of the target dataset, we construct a feature cloud for each target sample and leverage contrastive learning to extract more discriminative features. Lastly, we propose adaptive feature cloud expansion to push the decision boundary to a low-density area. Unlike most existing transfer learning methods that aim only to mitigate the domain gap, our method instead simultaneously considers the domain gap and the data deficiency problem of the target dataset. The conditional domain adaptation and the feature cloud generation of our method are learning jointly to extract discriminative features in an end-to-end manner. Besides, the adaptive feature cloud expansion improves the model's generalisation ability in the target domain. Extensive experiments on pneumonia and COVID-19 diagnosis tasks demonstrate that our method outperforms several state-of-the-art unsupervised domain adaptation approaches, which verifies the effectiveness of CDACM for automated pneumonia diagnosis using chest X-ray imaging.

Plasma gp96 is a Novel Predictive Biomarker for Severe COVID-19.

Early and effective identification of severe coronavirus disease 2019 (COVID-19) may allow us to improve the outcomes of associated severe acute respiratory illness with fever and respiratory symptoms. This study analyzed plasma concentrations of heat shock protein gp96 in nonsevere (including mild and typical) and severe (including severe and critical) patients with COVID-19 to evaluate its potential as a predictive and prognostic biomarker for disease severity. Plasma gp96 levels that were positively correlated with interleukin-6 (IL-6) levels were significantly elevated in COVID-19 patients admitted to the hospital but not in non-COVID-19 patients with less severe respiratory impairment. Meanwhile, significantly higher gp96 levels were observed in severe than nonsevere patients. Moreover, the continuous decline of plasma gp96 levels predicted disease remission and recovery, whereas its persistently high levels indicated poor prognosis in COVID-19 patients during hospitalization. Finally, monocytes were identified as the major IL-6 producers under exogenous gp96 stimulation. Our results demonstrate that plasma gp96 may be a useful predictive and prognostic biomarker for disease severity and outcome of COVID-19. IMPORTANCE Early and effective identification of severe COVID-19 may allow us to improve the outcomes of associated severe acute respiratory illness with fever and respiratory symptoms. Some heat shock proteins (Hsps) are released during oxidative stress, cytotoxic injury, and viral infection and behave as danger-associated molecular patterns (DAMPs). This study analyzed plasma concentrations of Hsp gp96 in nonsevere and severe patients with COVID-19. Significantly higher plasma gp96 levels were observed in severe than those in nonsevere patients, and its persistently high levels indicated poor prognosis in COVID-19 patients. The results demonstrate that plasma gp96 may be a useful predictive and prognostic biomarker for disease severity and outcome of COVID-19.

Clinical and pathological investigation of severe COVID-19 patients

BACKGROUND: Severe acute respiratory coronavirus 2 (SARS-CoV-2) caused coronavirus disease 2019 (COVID-19) has become a pandemic. This study addressed the clinical and immunopathological characteristics of severe COVID-19. METHODS: Sixty-nine COVID-19 patients were classified into as severe and non-severe groups to analyze their clinical and laboratory characteristics. A panel of blood cytokines was quantified over time. Biopsy specimens from two deceased cases were obtained for immunopathological, ultrastructural, and in situ hybridization examinations. RESULTS: Circulating cytokines, including IL8, IL6, TNFα, IP10, MCP1, and RANTES, were significantly elevated in severe COVID-19 patients. Dynamic IL6 and IL8 were associated with disease progression. SARS-CoV-2 was demonstrated to infect type II, type I pneumocytes and endothelial cells, leading to severe lung damage through cell pyroptosis and apoptosis. In severe cases, lymphopenia, neutrophilia, depletion of CD4+ and CD8+ T lymphocytes, and massive macrophage and neutrophil infiltrates were observed in both blood and lung tissues. CONCLUSIONS: A panel of circulating cytokines could be used to predict disease deterioration and inform clinical interventions. Severe pulmonary damage was predominantly attributed to both SARS-CoV-2 caused cytopathy and immunopathologic damage. Strategies that encourage pulmonary recruitment and overactivation of inflammatory cells by suppressing cytokine storm might improve the outcomes of severe COVID-19 patients.