Objective: To explore the value of contrast-enhanced computed tomography (CT) radiomics combined with machine learning algorithms in the preoperative prediction of perineural invasion (PNI) in oral squamous cell carcinoma (OSCC), aiming to provide evidence for assisting clinical treatment decision-making. Methods​: This study was approved by the Ethics Committee of the Ninth People's Hospital, Shanghai Jiao Tong University School of Medicine. A total of 250 OSCC patients confirmed by postoperative pathology were included, comprising 128 PNI-positive and 122 PNI-negative cases. The dataset was randomly divided into training (n=175), validation (n=38), and independent testing (n=37) sets in a ratio of 7:1.5:1.5. Regions of interest were delineated on preoperative images, and radiomic features were extracted. After dimensionality reduction and feature selection using methods like Least Absolute Shrinkage and Selection Operator (LASSO) regression, multiple machine learning models, including support vector machine (SVM), random forest, Light gradient boosting machine (LightGBM), and a Stacking ensemble model, were constructed. Model performance was evaluated using metrics such as the area under the receiver operating characteristic curve (AUC), sensitivity, calibration curves, and decision curve analysis (DCA). Model interpretability was analyzed using Shapley additive explanations (SHAP) and grouped permutation feature importance analysis. Results ​: ​ Among the 250 samples analyzed, the LightGBM model based on radiomics demonstrated the best performance on the independent test set, with an AUC of 0.781, outperforming models like SVM (AUC = 0.730) and Random Forest (AUC = 0.691), as well as clinical models (AUCs ranging 0.549-0.711). The LightGBM model showed good calibration (Brier score 0.198), and DCA indicated high clinical net benefit over a wide threshold probability range. Paired DeLong tests revealed no statistically significant differences in AUC between the ensemble (Stacking) model and the corresponding best-performing radiomics-based model. SHAP analysis and grouped permutation feature importance analysis further indicated that the primary discriminative information for the model came from radiomic texture features. Conclusion ​ The LightGBM model based on contrast-enhanced CT radiomics demonstrated good discriminative ability for preoperative prediction of PNI in OSCC. In the independent test set, it achieved the highest AUC. This model holds promise as a non-invasive auxiliary tool for preoperative risk assessment. Given the limited sample size of the independent test set, these results require further validation in larger cohorts and external datasets.

ObjectiveTo investigate the epidemiological and clinical characteristics of human bocavirus (HBoV) in hospitalized children with acute lower respiratory tract infection (ALRTI) at a single-center children’s hospital in Shanghai, thereby providing evidence for the diagnosis, treatment, and prevention of HBoV infection. MethodsA retrospective study was conducted on 19 537 hospitalized children with ALRTI at Shanghai Children’s Hospital from January 2021 to December 2023. Multiplex polymerase chain reaction (PCR) combined with capillary electrophoresis was used to detect HBoV and 12 other common respiratory viruses /atypical pathogens. The positive detection rate, demographic characteristics (sex, age), temporal distribution (year, season) of HBoV, as well as the clinical characteristics of severe and non-severe pneumonia were analyzed. ResultsThe overall HBoV-positive rate was 2.57% (503/19 537), with 59.44% (299/503) being single infections and 40.56% (204/503) being co-infections. The positive detection rate was significantly higher in boys than that in girls (2.78% vs 2.33%, χ²=3.88, P=0.049). The highest infection rate was observed in toddlers, followed by infants (χ²=379.57, P<0.001). The positive rate peaked in 2021 and reached its lowest point in 2023 (χ²=45.49, P<0.001), with epidemics mainly prevalent in summer and autumn. The main clinical symptoms were cough (90.06%, 453/503), fever (75.94%, 382/503), and wheezing (39.96%, 201/503). Children with severe pneumonia showed a higher incidence of wheezing compared with the non-severe group (P<0.001), while underlying diseases and co-infections had no significant association with disease severity (P>0.05). ConclusionHBoV was an important pathogen of ALRTI in children, predominantly affecting infants and toddlers, with higher susceptibility in boys and seasonal peaks in autumn and summer. The main clinical manifestations included cough, fever, and wheezing, with wheezing being more prevalent in children with severe pneumonia.

To compare the performance differences between the multidisciplinary team (MDT) model and the conventional diagnostic and treatment model for lung cancer, and to explore a high-quality development pathway for optimizing lung cancer diagnostic and treatment resources.

Bone metastases are a common and serious complication of solid tumors such as breast cancer, prostate cancer, lung cancer, and kidney cancer, and occur at a high rate in patients with advanced cancer. This pathological process not only frequently leads to skeletal-related events such as severe bone pain, pathological fractures, spinal cord compression, and hypercalcemia—conditions that severely impact patients’ quality of life—but may also significantly increase the risk of death. Cancer bone metastasis is a complex, multistage cascade involving key steps such as the detachment of tumor cells from the primary tumor, their invasion and entry into the bloodstream, and their colonization, dormancy, and reactivation within the bone microenvironment. In recent years, exercise—as a safe and easily implementable non-pharmacological intervention—has demonstrated significant potential in inhibiting bone metastasis, improving the bone microenvironment, and alleviating related clinical symptoms. Its mechanisms of action may involve multiple levels. First, at the mechanical load level, mechanical signals generated by exercise can directly act on bone cells, inhibiting tumor cell infiltration and colonization by regulating endothelial cell permeability and osteoclast activity. Second, at the endocrine and paracrine levels, exercise can alter the expression profiles of microRNAs (such as miR-486 and miR-34b) carried by extracellular vesicles and particles (EVPs) in the circulation, thereby inhibiting tumor cell proliferation and migration by targeting cell cycle-related genes. Furthermore, exercise can remodel the immune microenvironment, enhance the cytotoxic activity of antitumor immune cells, and improve oxygen supply to tumor tissues, thereby alleviating hypoxia-induced immunosuppression. Finally, at the metabolic intervention level, exercise can induce systemic metabolic reprogramming, increasing the nutritional demands of normal tissues, thereby competing with tumor cells for nutrients and forming a “metabolic barrier” in distal organs such as bones. This article reviews the pathogenesis and classification of bone metastases, including osteolytic, osteogenic, and mixed types, and elaborates in detail on the complex processes by which tumor cells undergo colonization, dormancy, and reactivation within the bone microenvironment. Furthermore, based on a comprehensive analysis of human and animal studies on exercise interventions for bone metastases, this paper highlights the mechanisms by which exercise inhibits tumor formation, growth, and spread through the regulation of mechanical loading, extracellular vesicles and granules, as well as via immune and metabolic pathways, thereby interfering with bone metastasis. However, the array of exercise modalities and intensities available for patients grappling with the challenges of cancer-related bone metastases remains markedly constrained, and the formulation of tailored exercise prescriptions continues to be devoid of robust, evidence-based medical backing. This review aims to explore the effects of exercise on bone health in patients with bone metastases, analyze the factors and considerations that should be taken into account in exercise prescriptions, and provide new theoretical support and insights for developing personalized exercise programs for these patients and further advancing relevant clinical research. Future research directions should include the development of evidence-based, personalized exercise prescriptions and further clarification of the safety boundaries and best practice standards for exercise interventions, thereby promoting the advancement of relevant clinical research and ultimately improving patients’ clinical outcomes and quality of life.

OBJECTIVE To provide data support and theoretical reference for drug-regulatory authorities to implement targeted supervision over internet-based pharmaceutical sales.METHODS Using keywords such as “internet-based pharmaceutical sales” and “online drug sales”， the official websites of the People’s Government， drug-regulatory authorities and databases including PKULaw and others were searched， so as to collect typical cases and public disclosure information of administrative penalties related to internet-based pharmaceutical sales from January 2014 to June 2024. Guided by accident-causation theory， the text-mining method was adopted to identify risk factors from four dimensions： human unsafe behaviors， unsafe state of substances， environmental factors and safety management. Grey relational analysis was performed to quantitatively evaluate and screen key risk factors， and targeted governance suggestions were subsequently put forward.RESULTS A total of 617 administrative penalty cases were included. The number of cases increased significantly since 2018. The regional distribution of cases presented a characteristic of “high in the east and low in the west”. Violations were mainly concentrated in the business qualification examination link， and online pharmaceutical retail enterprises were the dominant law-breaking subjects. A risk identification system for internet-based pharmaceutical sales was constructed， covering 4 primary-level， 17 secondary-level and 56 tertiary-level risk factors. Through grey relational analysis， 37 key tertiary-level risk factors （most of which belonged to the safety-management dimension） including inadequate bill management and illegal drug-advertising content， as well as 12 key secondary-level risk factors such as drug-advertising violations and non-standard drug storage， were screened out.CONCLUSIONS Inadequate capabilities and weak compliance awareness of practitioners， illegal drug advertising， non-standard drug storage and transportation， and imperfect quality-management systems of online-drug-sales enterprises and third-party platforms are prominent safety risk factors for internet-based pharmaceutical sales. It is recommended to strengthen qualification examination and competency training for practitioners， standardize drug-information display and drug advertising， improve the management system of drug warehousing and logistics， and refine institutional requirements for key links， so as to improve the supervision efficiency of internet-based pharmaceutical sales.

Five saponins were isolated from the kernels of Momordica cochinchinensis, by macroporous resin, silica gel, ODS column chromatography, and semi preparative HPLC. Based on MS and NMR analysis, combining with alkaline hydrolysis and acid hydrolysis, their structures were identified as: gypsogenin-3-O-{β-D-galactopyranosyl (1→2)-[α-L-rhamnopyranosyl (1→3)]-β-D-glucuropyranonosyl}-28-O-β-D-xylopyranosyl (1→3)-[β-D-xylopyranosyl (1→4)]-α-L-rhamnopyranosyl (1→2)-β-D-fucopyranoside (1), quillaic acid-3-O-{β-D-galactopyranosyl (1→2)-[α-L-rhamnopyranosyl (1→3)]-β-D-glucuropyranonosyl}-28-O-β-D-xylopyranosyl (1→3)-[β-D-xylopyranosyl (1→4)]-α-L-rhamnopyranosyl (1→2)-β-D-fucopyranoside (2), gypsogenin-3-O-β-D-galactopyranosyl (1→2)-[α-L-rhamnopyranosyl (1→3)]-β-D-glucuropyranonoside sodium (3), quillaic acid-3-O-β-D-galactopyranosyl (1→2)-[α-L-rhamnopyranosyl (1→3)]-β-D-glucuropyranonoside sodium (4), 18α-quillaic acid-3-O-β-D-galactopyranosyl (1→2)-[α-L-rhamnopyranosyl (1→3)]-β-D-glucuropyranonoside (5). Compounds 1-5 were new compounds, and named as mubezhisides A, B, C, D, E, respectively. They all could obviously inhibited the growth of Candida albicans, C. parapsilosis, and C. tropicalis.

ObjectiveTo retrospectively analyze the epidemiological trends of birth defects in perinatal infants in Fengxian District, Shanghai from 2018 to 2022, so as to provide a scientific evidence for the formulation of related prevention and control strategies. MethodsBased on the data from the National Birth Defects Surveillance System, statistical analysis was conducted on the perinatal birth defects from monitored hospitals within the region from 2018 to 2022. ResultsFrom 2018 to 2022, a total of 20 870 perinatal infants delivered in the monitored hospitals in Fengxian District, with 472 cases with birth defects, showing a significant increase in the prevalence of birth defects (PRR=1.49, 95%CI: 1.39‒1.59). The risk of birth defects increased with maternal age, especially for advanced maternal age (PRR=1.58, 95%CI: 1.12‒2.25). Infants born to mothers with gestational diabetes had a higher prevalence of birth defects compared to those without gestational diabetes (PRR=1.99, 95%CI: 1.46‒2.70). Infants with birth defects were more likely to be born prematurely (PRR=2.07, 95%CI:1.56‒2.76). The top three types of birth defects were congenital heart disease (CHD), other anomalies of the external ear, and polydactyly. ConclusionThe prevalence of birth defects in Fengxian District monitored hospitals showed an upward trend from 2018 to 2022. Advanced maternal age and gestational diabetes were identified as risk factors for birth defects. CHD is the leading type of birth defect in Fengxian District over the five-year period. To reduce the prevalence of birth defects, it is crucial to implement comprehensive prevention and treatment measures for CHD.

Purpose: Approximately 50%-74% of patients with metastatic human epidermal growth factor receptor 2 (HER2)–positive breast cancer do not respond to trastuzumab, with 75% of treated patients experiencing disease progression within a year. The combination of pyrotinib and capecitabine has showed efficacy in these patients. This study evaluates the efficacy and safety of pyrotinib combined with metronomic vinorelbine for trastuzumab-pretreated HER2-positive advanced breast cancer patients. Materials and Methods: In this phase 2 trial, patients aged 18-75 years with HER2-positive advanced breast cancer who had previously failed trastuzumab treatment were enrolled to receive pyrotinib 400 mg daily in combination with vinorelbine 40mg thrice weekly. The primary endpoint was progression-free survival (PFS), while secondary endpoints included objective response rate (ORR), disease control rate (DCR), overall survival (OS), and safety. Results: From October 21, 2019, to January 21, 2022, 36 patients were enrolled and received at least one dose of study treatment. At the cutoff date, 20 experienced disease progression or death. With a median follow-up duration of 35 months, the median PFS was 13.5 months (95% confidence interval [CI], 8.3 to 18.5). With all patients evaluated, an ORR of 38.9% (95% CI, 23.1 to 56.5) and a DCR of 83.3% (95% CI, 67.2 to 93.6) were achieved. The median OS was not reached. Grade 3 adverse events (AEs) were observed in 17 patients, with diarrhea being the most common (27.8%), followed by vomiting (8.3%) and stomachache (5.6%). There were no grade 4/5 AEs. Conclusion Pyrotinib combined with metronomic vinorelbine showed promising efficacy and an acceptable safety profile in HER2-positive advanced breast cancer patients after trastuzumab failure.

Hepatocellular carcinoma (HCC) is a leading cause of cancer-associated death globally. Liver transplantation (LT) has emerged as a key treatment for patients with HCC, and the Milan criteria have been adopted as the cornerstone of the selection policy. To allow more patients to benefit from LT, a number of expanded criteria have been proposed, many of which use radiologic morphological characteristics with larger and more tumors as surrogates to predict outcomes. Other groups developed indices incorporating biological variables and dynamic markers of response to locoregional treatment. These expanded selection criteria achieved satisfactory results with limited liver supplies. In addition, a number of prognostic models have been developed using clinicopathological characteristics, imaging radiomics features, genetic data, and advanced techniques such as artificial intelligence. These models could improve prognostic estimation, establish surveillance strategies, and bolster long-term outcomes in patients with HCC. In this study, we reviewed the latest findings and achievements regarding the selection criteria and post-transplant prognostic models for LT in patients with HCC.