Wei Zhang

This study aimed to develop an explainable machine learning–based model to enable early prediction of incident myocardial injury during hospitalization among patients with severe fever with thrombocytopenia syndrome (SFTS). This multicenter retrospective cohort study included and analyzed clinical data from 1,088 patients with SFTS who were hospitalized for the first time across four medical institutions between May 2011 and October 2024. The dataset was randomly split into an internal training set and a test set at a 7:3 ratio. After candidate predictors were selected from baseline clinical characteristics and laboratory indices, nine machine learning models were developed and compared, and external validation was conducted using retrospective cohorts from two independent medical centers. SHapley Additive Explanations (SHAP) were used to interpret model predictions and quantify feature importance. The ten key predictors identified comprised five baseline clinical characteristics and five laboratory indices. The CatBoost model demonstrated robust predictive performance in the internal test set and across two external validation cohorts, with AUCs of 0.750 (95% CI 0.698–0.803), 0.704 (95% CI 0.555–0.853), and 0.729 (95% CI 0.573–0.886), respectively. SHAP analysis further identified clinically relevant risk thresholds for the key laboratory predictors. In conclusion, this study developed a CatBoost‐based machine learning model that enables early prediction, at the time of hospital admission, of the risk of incident myocardial injury during hospitalization among patients with SFTS.

In China, there were few studies about the etiology, outcome and disease burden of severe viral encephalitis (VE) in recent years. The aims of this study w

Purpose: Using a multicenter clinical data, this study aimed to systematically analyze the clinical characteristics, pathogen spectrum, and independent risk/protective factors for coinfection in SFTS patients; additionally, it sought to explore optimal therapeutic regimens to reduce the case fatality rate of SFTS patients with coinfection. Methods: Clinical data of 1,675 patients with SFTS who were hospitalized for the first time, collected from 6 institutions between May 2011 and November 2024, were included and analyzed. Results: The coinfection group had a significantly higher case fatality rate (24.3% vs. 6.7%, P < 0.001), with the mixed bacterial-fungal infection subgroup showing the highest mortality risk (Log-rank test, P < 0.001). Albumin served as a protective factor against coinfection, with a cutoff value of 32 g/L identified; administering human albumin at this threshold could partially reduce the coinfection rate. Furthermore, on the basis of this protective effect of albumin, the combination of human albumin and corticosteroids was associated with a trend toward improved prognostic survival in coinfected patients. Conclusion: This study found a high proportion of SFTS patients with coinfection, which significantly worsens their clinical prognosis. SFTS currently has no specific therapy; clinical treatment focuses on symptomatic and supportive care, with strict standardization for antibiotics and antifungal use, and attention to patients’ nutritional support. Albumin is a key protective factor associated with a lower risk of coinfection in SFTS patients; combined use of human albumin and corticosteroids was associated with a further trend toward better clinical prognosis in coinfected patients.

Background: Viral encephalitis (VE), a central nervous system disorder with high mortality and disability rates, poses a serious threat to childhood development. Xiyanping injection (XYPI), an andrographolide sulfonate preparation widely used in China, exhibits anti-inflammatory, antiviral, antitumor, antibacterial and neuroprotective properties. Methods: We conducted a retrospective study of 635 pediatric inpatients with VE who were hospitalized at seven medical centers in China between January 2015 and December 2021. Risk factors for poor prognosis were compared between inpatients treated with XYPI (n = 480) and those without XYPI treatment (n = 155). Propensity score matching was performed to reduce potential confounding. Clinical symptoms, hospitalization costs, complications and sequelae were evaluated simultaneously. Results: Multivariate Logistic regression identified XYPI treatment as an independent protective factor for poor prognosis (odds ratio [OR] = 0.251, 95% confidence interval [CI]: 0.113–0.559, p < 0.001). XYPI significantly shortened the duration of fever and headache, reduced hospitalization costs, and lower the incidences of respiratory infections, myocardial injury, and sequelae (all p < 0.05). Conclusion: Adjuvant XYPI therapy may improve clinical outcomes and reduce the economic burden in pediatric VE; however, randomized trials are warranted to validate these findings.

Viral respiratory infections (VRTIs) caused by influenza (Flu) and COVID-19 pose significant global health challenges. Clinical outcomes are further exacerbated by infections with hospital acquired drug resistant pathogens (DRPs). A retrospective analysis was conducted on the data of 1,051 hospitalized patients with VRTIs from 2018 to 2024 at Beijing Ditan Hospital. Firstly, 280 drug-resistant strains were isolated from 185 patients with hospital-acquired DRPs infections for extended antibiogram analysis. Secondly, Interpretable machine learning (ML) was employed to predict the risk factors for hospital acquired DRPs infections in patients with VRTIs. Using the optimal feature subset, seven ML prediction models were developed. Parameter tuning was performed via 10-fold cross-validation and grid search. Model performance was evaluated using area under the curve (AUC), sensitivity, specificity, precision, and F1 score. SHAP (SHapley Additive exPlanations) was used to interpret the optimal model. Pathogen distribution in 280 clinical samples revealed sputum (65.36%) as themainsource, followed byblood (15.36%), urine (11.43%), and lavage fluid (5.00%). In all clinical specimens, Pseudomonas aeruginosa, Staphylococcus hominis, Escherichia coli, and Acinetobacter baumannii predominated in sputum, blood, urine, and lavage fluid, respectively. In terms of overall detection counts, the most frequently isolated strains were P. aeruginosa, Klebsiella pneumoniae, and A. baumannii. The drug resistance rate of P. aeruginosa to third-generation cephalosporins (such as ceftriaxone and cefotaxime) exceeds 89%, but it has relatively higher sensitivity to ceftazidime (71.7%) and cefepime (69.6%). Its drug resistance rates to imipenem and meropenem reach 45.7%. Although amikacin shows 100% sensitivity, combination with β-lactam antibiotics is recommended to reduce mortality. K. pneumoniae shows resistance rates of 53.3% to imipenem and 46.7% to meropenem, with over 50% resistance to levofloxacin and ciprofloxacin. Effective agents include sulfamethoxazole (68.9% susceptible), tigecycline (64.4%), chloramphenicol (62.2%), and amikacin (62.2%). Tigecycline combined with aminoglycosides has synergistic effects and inhibits resistant strains. A. baumannii was highly resistant to nearly all tested antibiotics, showing only partial susceptibility to minocycline (59.5%) and trimethoprim-sulfamethoxazole (38.1%). Among the seven ML models, the neural network (NN) achieved the best predictive performance. The SHAP method revealed the top 15 predictive variables by importance ranking, including length of stay (LOS), cholinesterase (CHE), age, albumin (ALB), etc.

Severe fever with thrombocytopenia syndrome (SFTS) is an emerging infectious disease with significant mortality risks. This study evaluated the predictive value of the lactate dehydrogenase/albumin to urea ratio (LAU) for lethal outcomes in SFTS patients. Clinical data from 348 patients (295 survivors, 53 non‐survivors) admitted to Yantai Infectious Disease Hospital (2022‐2023) were analyzed. Multivariate analysis identified older age, SFTS‐associated encephalopathy, SFTSV RNA, decreased albumin and serum calcium, increased lactate dehydrogenase, and elevated LAU as independent mortality risk factors. LAU demonstrated strong predictive accuracy (AUC: 0.79; 95% CI: 0.72‐0.85; sensitivity 69%, specificity 75%), with an optimal cutoff of 144.685 U·mmol/g·L. Survival analysis linked high LAU to poor prognosis, and longitudinal trends showed decreasing LAU with recovery in survivors but rising levels in non‐survivors. These findings highlight LAU as a novel, practical biomarker for early risk stratification and prognosis improvement in SFTS.

Severe fever with thrombocytopenia syndrome (SFTS) is a highly fatal disease. Droplet digital polymerase chain reaction (ddPCR) presents unparalleled sensitivity and enables absolute quantification of viral load. In this prospective study, we enrolled 111 patients with SFTS and collected 259 continuous samples. Our findings unveil a robust reverse transcription (RT)–ddPCR method for SFTS with a limit of detection of 2.46 copies/µL (95% CI, 1.50–11.05), surpassing the sensitivity of RT–quantitative polymerase chain reaction at 103.29 copies/µL (95% CI, 79.69–216.35). Longitudinal cohort analysis revealed significantly higher RT-ddPCR detection rates at days 10 to 11, 13 to 14, and ≥15 of the disease course as compared with RT–quantitative polymerase chain reaction (P < .05). Positive RT-ddPCR results were associated with declined platelet and elevated aspartate aminotransferase and lactate dehydrogenase on the same day vs negative RT-ddPCR samples. RT-ddPCR exhibits commendable diagnostic efficacy in SFTS, and it remains detectable in blood samples from patients with an extended disease course. Furthermore, RT-ddPCR correlates with clinical laboratory tests, furnishing valuable reference data for clinical diagnosis.

Severe fever with thrombocytopenia syndrome (SFTS) is an acute infectious disease prevalent in East Asia with a high mortality rate (5%–30%). Reverse transcription loop‐mediated isothermal amplification (RT‐LAMP), a rapid nucleic acid‐based diagnostic technique, is a useful alternative for the clinical diagnosis of SFTS, particularly in resource‐limited hospitals or rural clinics in SFTS virus‐endemic regions. However, the actual clinical sensitivity and specificity of RT‐LAMP remain unclear. This study evaluated the field application of RT‐LAMP. This prospective field study included 130 patients with laboratory‐confirmed SFTS from Yantai, Shandong Province, China. Two sets of RT‐LAMP primers were validated, and one set of RT‐LAMP assays was optimized for field detection. Nucleic acids of serially collected serum/plasma samples were identified using quantitative reverse transcription polymerase chain reaction (RT‐qPCR) and RT‐LAMP. In laboratory tests, we optimized the detection time of primer set 2 for the RT‐LAMP to 60 min. Notably, the onsite testing of 279 plasma samples from patients with SFTS revealed that the sensitivity and specificity of the test were 81.9% and 96.3%, respectively. We also analyzed samples with different durations of the disease, and our study showed that the sensitivity of RT‐LAMP detection at the beginning of admission was 89.92%. Univariate analysis showed that the detection rate of RT‐LAMP was similar to that of RT‐qPCR in the first 5 days of the disease course and was lower than that of RT‐qPCR on Days 6 and 14–15 of the disease course. The positive detection rate in patients aged ≥ 65 years was significantly higher than that in younger age groups. RT‐LAMP is a simple, suitable, and rapid clinical detection method of SFTS onsite screening. It is more suitable for screening patients in the early stages of the disease and analyzing samples obtained from patients aged ≥ 65 years before the 6th day of the disease course.