MultiR-Net: A Novel Joint Learning Network for COVID-19 segmentation and classification.

The outbreak of COVID-19 has caused a severe shortage of healthcare resources. Ground Glass Opacity (GGO) and consolidation of chest CT scans have been an essential basis for imaging diagnosis since 2020. The similarity of imaging features between COVID-19 and other pneumonia makes it challenging to distinguish between them and affects radiologists' diagnosis. Recently, deep learning in COVID-19 has been mainly divided into disease classification and lesion segmentation, yet little work has focused on the feature correlation between the two tasks. To address these issues, in this study, we propose MultiR-Net, a 3D deep learning model for combined COVID-19 classification and lesion segmentation, to achieve real-time and interpretable COVID-19 chest CT diagnosis. Precisely, the proposed network consists of two subnets: a multi-scale feature fusion UNet-like subnet for lesion segmentation and a classification subnet for disease diagnosis. The features between the two subnets are fused by the reverse attention mechanism and the iterable training strategy. Meanwhile, we proposed a loss function to enhance the interaction between the two subnets. Individual metrics can not wholly reflect network effectiveness. Thus we quantify the segmentation results with various evaluation metrics such as average surface distance, volume Dice, and test on the dataset. We employ a dataset containing 275 3D CT scans for classifying COVID-19, Community-acquired Pneumonia (CAP), and healthy people and segmented lesions in pneumonia patients. We split the dataset into 70% and 30% for training and testing. Extensive experiments showed that our multi-task model framework obtained an average recall of 93.323%, an average precision of 94.005% on the classification test set, and a 69.95% Volume Dice score on the segmentation test set of our dataset.

Assessment of a biofluid mechanics-based model for calculating portal pressure in canines.

BACKGROUND: Portal hypertension is a severe complication caused by various chronic liver diseases. The standard methods for detecting portal hypertension (hepatic venous pressure gradient and free portal pressure) are available in only a few hospitals due to their technical difficulty and invasiveness; thus, non-invasive measuring methods are needed. This study aimed to establish and assess a novel model to calculate free portal pressure based on biofluid mechanics. RESULT: Comparison of each dog's virtual and actual free portal pressure showed that a biofluid mechanics-based model could accurately predict free portal pressure (mean difference: -0.220, 95% CI: - 0.738 to 0.298; upper limit of agreement: 2.24, 95% CI: 1.34 to 3.14; lower limit of agreement: -2.68, 95% CI: - 3.58 to - 1.78; intraclass correlation coefficient: 0.98, 95% CI: 0.96 to 0.99; concordance correlation coefficient: 0.97, 95% CI: 0.93 to 0.99) and had a high AUC (0.984, 95% CI: 0.834 to 1.000), sensitivity (92.3, 95% CI: 64.0 to 99.8), specificity (91.7, 95% CI: 61.5 to 99.8), positive likelihood ratio (11.1, 95% CI: 1.7 to 72.8), and low negative likelihood ratio (0.08, 95% CI: 0.01 to 0.6) for detecting portal hypertension. CONCLUSIONS: Our study suggests that the biofluid mechanics-based model was able to accurately predict free portal pressure and detect portal hypertension in canines. With further research and validation, this model might be applicable for calculating human portal pressure, detecting portal hypertensive patients, and evaluating disease progression and treatment efficacy.