Objective To investigate the value of early cystatin C(Cys-C)combined with prothrombin time(PT)for assessing the severity and outcome of patients with severe fever with thrombocytopenia syndrome(SFTS).Methods The data of 101 patients with SFTS diagnosed and treated in Chaohu Hospital of Anhui Medical University from April 2021 to August 2023 were reviewed retrospectively.The patients were assigned to non-severe group or severe group according to the severity of the disease,and assigned to survivors group or deaths group according to the treatment outcome.The clinical manifestations and early laboratory test results were compared between groups.Logistic regression model was used to analyze the factors for predicting the outcome.The receiver operating characteristic(ROC)curve was used to evaluate the ability of Cys-C and PT levels alone and in combination to distinguish survivors from deaths.Results In the early stage,the patients in severe group and deaths group showed significantly higher levels of serum Cys-C,blood urea nitrogen(BUN),creatinine(Cr),uric acid(UA),PT,activated partial thromboplastin time(APTT),creatine kinase(CK)and creatine kinase MB(CKMB)but significantly lower lymphocyte count compared to the patients in the non-severe group and survivors group(P＜0.05).Multivariate binary Logistic regression analysis showed that age(OR=1.146,95％CI:1.036-1.267),PT(OR=2.643,95％CI:1.323-5.281),and Cys-C(OR=5.039,95％CI:1.548-16.395)were independent risk factors for the outcome of SFTS patients.Serum Cys-C was valuable in distinguishing survivors from deaths,the area under the ROC curve of which was 0.831.When Cys-C,PT and age were combined,the AUC was up to 0.930.Conclusions SFTS mainly occurs in farmers and elderly people.Serum PT and Cys-C levels at early stage can effectively predict the outcome of patients with SFTS.PT and Cys-C levels are expected to be biomarkers for distinguishing survivors from deaths.

Background/Aims: The accuracy of endosonographers in diagnosing gastric subepithelial lesions (SELs) using endoscopic ultrasonography (EUS) is influenced by experience and subjectivity. Artificial intelligence (AI) has achieved remarkable development in this field. This study aimed to develop an AI-based EUS diagnostic model for the diagnosis of SELs, and evaluated its efficacy with external validation. Methods: We developed the EUS-AI model with ResNeSt50 using EUS images from two hospitals to predict the histopathology of the gastric SELs originating from muscularis propria. The diagnostic performance of the model was also validated using EUS images obtained from four other hospitals. Results: A total of 2,057 images from 367 patients (375 SELs) were chosen to build the models, and 914 images from 106 patients (108 SELs) were chosen for external validation. The sensitivity, specificity, positive predictive value, negative predictive value, and accuracy of the model for differentiating gastrointestinal stromal tumors (GISTs) and non-GISTs in the external validation sets by images were 82.01%, 68.22%, 86.77%, 59.86%, and 78.12%, respectively. The sensitivity, specificity, positive predictive value, negative predictive value, and accuracy in the external validation set by tumors were 83.75%, 71.43%, 89.33%, 60.61%, and 80.56%, respectively. The EUS-AI model showed better performance (especially specificity) than some endosonographers.The model helped improve the sensitivity, specificity, and accuracy of certain endosonographers. Conclusions We developed an EUS-AI model to classify gastric SELs originating from muscularis propria into GISTs and non-GISTs with good accuracy. The model may help improve the diagnostic performance of endosonographers. Further work is required to develop a multi-modal EUS-AI system.