https://www.selleckchem.com/products/enarodustat.html 7%) along the bronchial vascular bundle. Five cases (83.3%) had ground-glass opacities, 4 cases (66.7%) had ground-glass nodules, 1 case (16.7%) had thickened lobular septum, 2 cases (33.3%) had thickened bronchial wall, 2 cases (33.3%) had halo sign,1 case (16.7%) had crazy-paving sign, and 1 case (16.7%) had tree-in-bud sign. CONCLUSIONS The imaging manifestations of early-stage COVID-19 are relatively mild, and the imaging findings of some patients are not typical, which can easily lead to missed diagnoses. Thus, suspected cases need to be closely monitored, and epidemiological history and clinical laboratory examination should also be considered during diagnosis.Recurrence is a major cause of cancer-related deaths in colorectal cancer (CRC) patients, but the current strategies are limited to predict this clinical behavior. Our aim is to develop a recurrence prediction model based on long non-coding RNAs (lncRNAs) in exosomes of serum to improve the prediction accuracy. In discovery phase, 11 lncRNAs were found to be associated with CRC recurrence in tissues using high-throughput lncRNAs microarray and reverse transcription quantitative real-time PCR. And, 9 of them were correlated with their expression levels of serum exosomes. In training phase, a model based on 5-exosomal lncRNAs (exolncRNAs) panel was constructed, and showed high distinguish capability for recurrent CRC patients. ROC showed the panel was superior to serum CEA and CA19-9 in prediction of CRC recurrence. In both training and test sets, high-risk patients defined by the 5-exolncRNAs panel had poor recurrence free and overall survival. And, COX model showed it was an independent factor for CRC prognosis. Moreover, there was a significant relationship in detection of 5-exolncRNAs between plasma samples and paired serum samples. In summary, the 5-exolncRNAs panel robustly stratifies CRC patients' risk of recurrence, enabling more accurate predic