Medical-VLBERT: Medical Visual Language BERT for COVID-19 CT Report Generation With Alternate Learning.

Medical imaging technologies, including computed tomography (CT) or chest X-Ray (CXR), are largely employed to facilitate the diagnosis of the COVID-19. Since manual report writing is usually too time-consuming, a more intelligent auxiliary medical system that could generate medical reports automatically and immediately is urgently needed. In this article, we propose to use the medical visual language BERT (Medical-VLBERT) model to identify the abnormality on the COVID-19 scans and generate the medical report automatically based on the detected lesion regions. To produce more accurate medical reports and minimize the visual-and-linguistic differences, this model adopts an alternate learning strategy with two procedures that are knowledge pretraining and transferring. To be more precise, the knowledge pretraining procedure is to memorize the knowledge from medical texts, while the transferring procedure is to utilize the acquired knowledge for professional medical sentences generations through observations of medical images. In practice, for automatic medical report generation on the COVID-19 cases, we constructed a dataset of 368 medical findings in Chinese and 1104 chest CT scans from The First Affiliated Hospital of Jinan University, Guangzhou, China, and The Fifth Affiliated Hospital of Sun Yat-sen University, Zhuhai, China. Besides, to alleviate the insufficiency of the COVID-19 training samples, our model was first trained on the large-scale Chinese CX-CHR dataset and then transferred to the COVID-19 CT dataset for further fine-tuning. The experimental results showed that Medical-VLBERT achieved state-of-the-art performances on terminology prediction and report generation with the Chinese COVID-19 CT dataset and the CX-CHR dataset. The Chinese COVID-19 CT dataset is available at https://covid19ct.github.io/.

Comparison of Clinical Features and CT Temporal Changes Between Familial Clusters and Non-familial Patients With COVID-19 Pneumonia.

Purpose: This study aimed to compare the clinical characteristics, laboratory findings, and chest computed tomography (CT) findings of familial cluster (FC) and non-familial (NF) patients with coronavirus disease 2019 (COVID-19) pneumonia. Methods: This retrospective study included 178 symptomatic adult patients with laboratory-confirmed COVID-19. The 178 patients were divided into FC (n = 108) and NF (n = 70) groups. Patients with at least two confirmed COVID-19 cases in their household were classified into the FC group. The clinical and laboratory features between the two groups were compared and so were the chest CT findings on-admission and end-hospitalization. Results: Compared with the NF group, the FC group had a longer period of exposure (13.1 vs. 8.9 days, p < 0.001), viral shedding (21.5 vs. 15.9 days, p < 0.001), and hospital stay (39.2 vs. 22.2 days, p < 0.001). The FC group showed a higher number of involved lung lobes on admission (3.0 vs. 2.3, p = 0.017) and at end-hospitalization (3.6 vs. 1.7, p < 0.001) as well as higher sum severity CT scores at end-hospitalization (4.6 vs. 2.7, p = 0.005) than did the NF group. Conversely, the FC group had a lower lymphocyte count level (p < 0.001) and a significantly lower difference in the number of involved lung lobes (Δnumber) between admission and discharge (p < 0.001). Notably, more cases of severe or critical illness were observed in the FC group than in the NF group (p = 0.036). Conclusions: Patients in the FC group had a worse clinical course and outcome than those in the NF group; thus, close monitoring during treatment and follow-ups after discharge would be beneficial for patients with familial infections.

CT-based radiomics for predicting the rapid progression of coronavirus disease 2019 (COVID-19) pneumonia lesions.

OBJECTIVES: To develop and validate a radiomic model to predict the rapid progression (defined as volume growth of pneumonia lesions > 50% within seven days) in patients with coronavirus disease 2019 (COVID-19). METHODS: Patients with laboratory-confirmed COVID-19 who underwent longitudinal chest CT between January 01 and February 18, 2020 were included. A total of 1316 radiomic features were extracted from the lung parenchyma window for each CT. The least absolute shrinkage and selection operator (LASSO), Relief, Las Vegas Wrapper (LVW), L1-norm-Support Vector Machine (L1-norm-SVM), and recursive feature elimination (RFE) were applied to select the features that associated with rapid progression. Four machine learning classifiers were used for modeling, including Support Vector Machine (SVM), Random Forest (RF), Logistic Regression (LR), and Decision Tree (DT). Accordingly, 20 radiomic models were developed on the basis of 296 CT scans and validated in 74 CT scans. Model performance was determined by the receiver operating characteristic curve. RESULTS: A total of 107 patients (median age, 49.0 years, interquartile range, 35-54) were evaluated. The patients underwent a total of 370 chest CT scans with a median interval of 4 days (interquartile range, 3-5 days). The combination methods of L1-norm SVM and SVM with 17 radiomic features yielded the highest performance in predicting the likelihood of rapid progression of pneumonia lesions on next CT scan, with an AUC of 0.857 (95% CI: 0.766-0.947), sensitivity of 87.5%, and specificity of 70.7%. CONCLUSIONS: Our radiomic model based on longitudinal chest CT data could predict the rapid progression of pneumonia lesions, which may facilitate the CT follow-up intervals and reduce the radiation. ADVANCES IN KNOWLEDGE: Radiomic features extracted from the current chest CT have potential in predicting the likelihood of rapid progression of pneumonia lesions on the next chest CT, which would improve clinical decision-making regarding timely treatment.

Three-Dimensional CT for Quantification of Longitudinal Lung and Pneumonia Variations in COVID-19 Patients.

Objectives: Visual chest CT is subjective with interobserver variability. We aimed to quantify the dynamic changes of lung and pneumonia on three-dimensional CT (3D-CT) images in coronavirus disease 2019 (COVID-19) patients during hospitalization. Methods: A total of 110 laboratory-confirmed COVID-19 patients who underwent chest CT from January 3 to February 29, 2020 were retrospectively reviewed. Pneumonia lesions were classified as four stages: early, progressive, peak, and absorption stages on chest CT. A computer-aided diagnostic (CAD) system calculated the total lung volume (TLV), the percentage of low attenuation areas (LAA%), the volume of pneumonia, the volume of ground-glass opacities (GGO), the volume of consolidation plus the GGO/consolidation ratio. The CT score was visually assessed by radiologists. Comparisons of lung and pneumonia parameters among the four stages were performed by one-way ANOVA with post-hoc tests. The relationship between the CT score and the volume of pneumonia, and between LAA% and the volume of pneumonia in four stages was assessed by Spearman's rank correlation analysis. Results: A total of 534 chest CT scans were performed with a median interval of 4 days. TLV, LAA%, and the GGO/consolidation ratio were significantly decreased, while the volume of pneumonia, GGO, and consolidation were significantly increased in the progressive and peak stages (for all, P < 0.05). The CT score was significantly correlated with the pneumonia volume in the four stages (r = 0.731, 0.761, 0.715, and 0.669, respectively, P < 0.001). Conclusion: 3D-CT could be used as a useful quantification method in monitoring the dynamic changes of COVID-19 pneumonia.

Machine learning models for predicting critical illness risk in hospitalized patients with COVID-19 pneumonia.

BACKGROUND: To develop machine learning classifiers at admission for predicting which patients with coronavirus disease 2019 (COVID-19) who will progress to critical illness. METHODS: A total of 158 patients with laboratory-confirmed COVID-19 admitted to three designated hospitals between December 31, 2019 and March 31, 2020 were retrospectively collected. 27 clinical and laboratory variables of COVID-19 patients were collected from the medical records. A total of 201 quantitative CT features of COVID-19 pneumonia were extracted by using an artificial intelligence software. The critically ill cases were defined according to the COVID-19 guidelines. The least absolute shrinkage and selection operator (LASSO) logistic regression was used to select the predictors of critical illness from clinical and radiological features, respectively. Accordingly, we developed clinical and radiological models using the following machine learning classifiers, including naive bayes (NB), linear regression (LR), random forest (RF), extreme gradient boosting (XGBoost), adaptive boosting (AdaBoost), K-nearest neighbor (KNN), kernel support vector machine (k-SVM), and back propagation neural networks (BPNN). The combined model incorporating the selected clinical and radiological factors was also developed using the eight above-mentioned classifiers. The predictive efficiency of the models is validated using a 5-fold cross-validation method. The performance of the models was compared by the area under the receiver operating characteristic curve (AUC). RESULTS: The mean age of all patients was 58.9±13.9 years and 89 (56.3%) were males. 35 (22.2%) patients deteriorated to critical illness. After LASSO analysis, four clinical features including lymphocyte percentage, lactic dehydrogenase, neutrophil count, and D-dimer and four quantitative CT features were selected. The XGBoost-based clinical model yielded the highest AUC of 0.960 [95% confidence interval (CI): 0.913-1.000)]. The XGBoost-based radiological model achieved an AUC of 0.890 (95% CI: 0.757-1.000). However, the predictive efficacy of XGBoost-based combined model was very close to that of the XGBoost-based clinical model, with an AUC of 0.955 (95% CI: 0.906-1.000). CONCLUSIONS: A XGBoost-based based clinical model on admission might be used as an effective tool to identify patients at high risk of critical illness.

Clinical Utility of a Nomogram for Predicting 30-Days Poor Outcome in Hospitalized Patients With COVID-19: Multicenter External Validation and Decision Curve Analysis.

Aim: Early detection of coronavirus disease 2019 (COVID-19) patients who are likely to develop worse outcomes is of great importance, which may help select patients at risk of rapid deterioration who should require high-level monitoring and more aggressive treatment. We aimed to develop and validate a nomogram for predicting 30-days poor outcome of patients with COVID-19. Methods: The prediction model was developed in a primary cohort consisting of 233 patients with laboratory-confirmed COVID-19, and data were collected from January 3 to March 20, 2020. We identified and integrated significant prognostic factors for 30-days poor outcome to construct a nomogram. The model was subjected to internal validation and to external validation with two separate cohorts of 110 and 118 cases, respectively. The performance of the nomogram was assessed with respect to its predictive accuracy, discriminative ability, and clinical usefulness. Results: In the primary cohort, the mean age of patients was 55.4 years and 129 (55.4%) were male. Prognostic factors contained in the clinical nomogram were age, lactic dehydrogenase, aspartate aminotransferase, prothrombin time, serum creatinine, serum sodium, fasting blood glucose, and D-dimer. The model was externally validated in two cohorts achieving an AUC of 0.946 and 0.878, sensitivity of 100 and 79%, and specificity of 76.5 and 83.8%, respectively. Although adding CT score to the clinical nomogram (clinical-CT nomogram) did not yield better predictive performance, decision curve analysis showed that the clinical-CT nomogram provided better clinical utility than the clinical nomogram. Conclusions: We established and validated a nomogram that can provide an individual prediction of 30-days poor outcome for COVID-19 patients. This practical prognostic model may help clinicians in decision making and reduce mortality.

Treatment With Convalescent Plasma for Critically Ill Patients With Severe Acute Respiratory Syndrome Coronavirus 2 Infection.

As of March 24, 2020, novel severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has been responsible for 379,661 infection cases with 16,428 deaths globally, and the number is still increasing rapidly. Herein, we present four critically ill patients with SARS-CoV-2 infection who received supportive care and convalescent plasma. Although all four patients (including a pregnant woman) recovered from SARS-CoV-2 infection eventually, randomized trials are needed to eliminate the effect of other treatments and investigate the safety and efficacy of convalescent plasma therapy.

Novel coronavirus disease 2019 (COVID-19): relationship between chest CT scores and laboratory parameters.

PURPOSE: To quantify the severity of 2019 novel coronavirus disease (COVID-19) on chest CT and to determine its relationship with laboratory parameters. METHODS: Patients with real-time fluorescence polymerase chain reaction (RT-PCR)-confirmed COVID-19 between January 01 and February 18, 2020, were included in this study. Laboratory parameters were retrospectively collected from medical records. Severity of lung changes on chest CT of early, progressive, peak, and absorption stages was scored according to the percentage of lung involvement (5 lobes, scores 1-5 for each lobe, range 0-20). Relationship between CT scores and laboratory parameters was evaluated by the Spearman rank correlation. The Bonferroni correction adjusted significance level was at 0.05/4 = 0.0125. RESULTS: A total of 84 patients (mean age, 47.8 ± 12.0 years [standard deviation]; age range, 24-80 years) were evaluated. The patients underwent a total of 339 chest CT scans with a median interval of 4 days (interquartile range, 3-5 days). Median chest CT scores peaked at 4 days after the beginning of treatment and then declined. CT score of the early stage was correlated with neutrophil count (r = 0.531, P = 0.011). CT score of the progressive stage was correlated with neutrophil count (r = 0.502, P < 0.001), white blood cell count (r = 0.414, P = 0.001), C-reactive protein (r = 0.511, P < 0.001), procalcitonin (r = 0.423, P = 0.004), and lactose dehydrogenase (r = 0.369, P = 0.010). However, CT scores of the peak and absorption stages were not correlated with any parameter (P > 0.0125). No sex difference occurred regarding CT score (P > 0.05). CONCLUSION: Severity of lung abnormalities quantified on chest CT might correlate with laboratory parameters in the early and progressive stages. However, larger cohort studies are necessary.