A Multiclass Radiomics Method-Based WHO Severity Scale for Improving COVID-19 Patient Assessment and Disease Characterization From CT Scans.

OBJECTIVES: The aim of this study was to evaluate the severity of COVID-19 patients' disease by comparing a multiclass lung lesion model to a single-class lung lesion model and radiologists' assessments in chest computed tomography scans. MATERIALS AND METHODS: The proposed method, AssessNet-19, was developed in 2 stages in this retrospective study. Four COVID-19-induced tissue lesions were manually segmented to train a 2D-U-Net network for a multiclass segmentation task followed by extensive extraction of radiomic features from the lung lesions. LASSO regression was used to reduce the feature set, and the XGBoost algorithm was trained to classify disease severity based on the World Health Organization Clinical Progression Scale. The model was evaluated using 2 multicenter cohorts: a development cohort of 145 COVID-19-positive patients from 3 centers to train and test the severity prediction model using manually segmented lung lesions. In addition, an evaluation set of 90 COVID-19-positive patients was collected from 2 centers to evaluate AssessNet-19 in a fully automated fashion. RESULTS: AssessNet-19 achieved an F1-score of 0.76 ± 0.02 for severity classification in the evaluation set, which was superior to the 3 expert thoracic radiologists (F1 = 0.63 ± 0.02) and the single-class lesion segmentation model (F1 = 0.64 ± 0.02). In addition, AssessNet-19 automated multiclass lesion segmentation obtained a mean Dice score of 0.70 for ground-glass opacity, 0.68 for consolidation, 0.65 for pleural effusion, and 0.30 for band-like structures compared with ground truth. Moreover, it achieved a high agreement with radiologists for quantifying disease extent with Cohen κ of 0.94, 0.92, and 0.95. CONCLUSIONS: A novel artificial intelligence multiclass radiomics model including 4 lung lesions to assess disease severity based on the World Health Organization Clinical Progression Scale more accurately determines the severity of COVID-19 patients than a single-class model and radiologists' assessment.

Qualitative and quantitative analysis of diffusion-weighted brain MR imaging in comatose survivors after cardiac arrest.

PURPOSE: The aim of this study is to compare a qualitative and a quantitative assessment of brain diffusion-weighted imaging (DWI) in predicting outcome of comatose patients after cardiac arrest (CA). METHODS: Two observers used a scoring template to analyze the DWI of 75 patients. A total of 13 regions were scored from 0 to 3 (0 = normal, 1 = probably normal, 2 = probably abnormal, 3 = definitely abnormal). The total cerebral cortex (TCC), the total deep grey nuclei (TDGN), the total brain stem, the total cerebellum, and the total brain score were calculated. Intra- and inter-observer variability were tested. The mean whole brain apparent diffusion coefficient (ADC) values and percentage of voxels below a specific ADC value cut-off were calculated. The data were correlated with clinical outcome (cerebral performance category score after 180 days, dichotomized in a score 1-2 with favorable outcome and score 3-5 with unfavorable outcome) using ROC analysis. RESULTS: Intra-observer variability was excellent for the TCC score (ICC 0.95 and 0.86) and the TDGN score (ICC 0.89 and 0.75). Inter-observer variability was good to excellent for total cerebral cortex score and total deep grey nuclei score in both the first (ICC 0.78 and 0.69) and third (ICC 0.86 and 0.83) image assessment. TCC and TDGN score show the best correlation with clinical outcome (highest AUC values 0.87 and 0.87). Quantitative parameters did not show good correlation with clinical outcome (AUC values 0.57 and 0.60). CONCLUSION: A qualitative assessment of brain DWI using a scoring template provides useful data regarding patient outcome while quantitative data appeared less reliable.