Affiliated Hospital of Hangzhou Normal University

Metabolic dysfunction-associated steatotic liver disease (MASLD) can progress to metabolic dysfunction-associated steatohepatitis (MASH) and liver fibrosis, contributing to a heavier global health burden. Non-invasive diagnostic tools developed using machine learning (ML) and deep learning (DL), two representative artificial intelligence algorithms, are increasingly being explored for MASH and its related fibrosis assessment. This study aimed to compare the diagnostic performance of different ML and DL models and identify the top-performing models for diagnosing MASH and associated liver fibrosis. A systematic review and meta-analysis were conducted across PubMed, Web of Science, Embase and Cochrane Library from inception to May 18, 2025. Pooled area under the receiver operator characteristic curve (AUROC) values with 95 % confidence interval (CI) were calculated. Accuracy, specificity, sensitivity, positive predictive values, and negative predictive values were also recorded. Of 4,314 studies initially identified, 106 met the inclusion criteria, with 35 studies (ML: n=28; DL: n=7) providing data for analysis. Logistic Regression and Neural Network are the most commonly algorithms applied in ML and DL, respectively. The pooled AUROCs for diagnosing MASH were 0.833 (95 %CI: 0.806-0.860) for ML models and 0.841 (95 %CI: 0.782-0.900) for DL models. Light Gradient Boosting Machine (LightGBM) and ResNet50 were the best-performing models for diagnosing MASH within ML and DL algorithms, respectively, achieving corresponding AUROCs of 0.920 (95 %CI: 0.916-0.924) and 0.960 (95 %CI: 0.951-0.969). For fibrosis diagnosis, ML models had a pooled AUROC of 0.826 (95 %CI: 0.792-0.860), with Categorical Boosting (CatBoost) achieving the highest AUROC of 0.960 (95 %CI: 0.950-0.970). DL models yielded the pooled AUROC of 0.875 (95 %CI: 0.816-0.934) for fibrosis diagnosis. Both ML and DL models demonstrated strong diagnostic performance for MASH and liver fibrosis, with DL achieving marginally higher AUROCs. AI-driven approaches show promise in MASLD management.

Multiple sclerosis (MS) is a progressive, long-term disorder affecting the central nervous system. The management and treatment of MS require significant medical resources, placing a heavy burden on both individuals and society. Autophagy is essential for the degradation of dysfunctional or excess cellular components. In the context of MS, autophagy exhibits dual roles, both protective and detrimental. On one hand, it mitigates disease progression by reducing oxidative stress and inflammation. On the other hand, autophagy activates various immune and supportive cells pivotal in MS pathogenesis. This review aims to explore the relationship between autophagy and MS, its impact on disease progression and the current challenges in this field.

Against the backdrop of normalized COVID-19 prevention and control, this study focuses on community health service centers. Here, “Cultural values” refer to the shared beliefs, norms, and attitudes within a community or organization that shape behavior and decision-making, such as a sense of collective responsibility, mutual support, and adherence to public health guidelines. This study explores the impact of Cultural values and social capital on their COVID-19 response capabilities and proposes targeted improvement strategies. A self-designed questionnaire was used to survey 983 medical staff from 15 community health service centers in each of six cities in Zhejiang Province. Differences in COVID-19 response capabilities across centers with varying demographic characteristics were compared using t-tests and variance analysis. Hierarchical linear regression was employed to analyze influencing factors, while structural equation modeling path analysis explored the impact pathways and mechanisms of Cultural values and social capital on COVID-19 response capabilities. Cultural values have a significant impact on the COVID-19 response capability. The overall effect value, direct effect value, and indirect effect value of cultural appeal on the COVID-19 response capability of community health service centers are 0.784, 0.070, and 0.070 respectively. Social capital exhibits a partial mediating effect. Among them, the overall effect value of structural social capital is 0.094, that of relational social capital is 0.260, and that of cognitive social capital is 0.241. The influence of cultural appeal on pandemic response capability is mainly transmitted indirectly through relational and cognitive social capital, while the mediating effect of structural social capital is not significant. Cultural values and social capital positively influence COVID-19 response capabilities in community health service centers, with Cultural values indirectly affecting these capabilities through relational and cognitive social capital. Thus, centers can enhance their COVID-19 response capabilities by strengthening Cultural values, fostering social capital, integrating health resources, and establishing collaborative mechanisms, ensuring community residents’ health and safety during crises like sudden outbreaks and natural disasters.

Plant-derived self-assembled nanoparticles, especially from food-medicine homology sources, are gaining attention, yet their structure-function relationships remain unclear. This study identified such nanoparticles from leaf decoction of Eucommia ulmoides, a key plant in traditional Asian medicine and diet, termed EUPs. These spherical particles (~287.8 nm) were primarily composed of polysaccharides and polyphenols, with 268 polyphenolic compounds detected via UPLC-QTOF-MS. Stepwise dissociation-ultrafiltration and spectroscopic analyses revealed that polyphenols were bound to the polysaccharide through noncovalent interactions, including hydrogen bonding and hydrophobic forces, forming a layered structure with sustained-release and thermo-responsive properties. Compared with free polyphenols, EUPs exhibited significantly prolonged anti-inflammatory effects in LPS-stimulated RAW 264.7 macrophages, reflected by the suppression of key inflammatory cytokines, including TNF-α, IL-6 and NO. Therefore, it aims to offer mechanistic insights into the multi-component synergistic anti-inflammatory effects of E. ulmoides and supporting the material basis of food-medicine homology.

BackgroundNon-small cell lung cancer (NSCLC) accounts for 80–85% of lung cancers and remains the leading cause of cancer-related mortality worldwide. Immune checkpoint inhibitors (ICIs) improve survival in selected patients, yet their clinical utility is limited by low objective response rates, challenges in patient selection, and suboptimal performance of existing biomarkers. Radiomics can quantitatively characterize tumor heterogeneity on computed tomography (CT) imaging and shows promise in predicting treatment response and immune microenvironment profiles. CD38, a key immunosuppressive molecule, impairs T-cell function and recruits suppressive cells via the adenosine pathway. This study aimed to integrate radiomic features with CD38 expression to build a multimodal prediction model and investigate its underlying biology.MethodsWe retrospectively included 45 NSCLC patients receiving ICIs (training cohort: n=31; validation cohort: n=14). A total of 1,223 CT radiomic features were extracted and selected by least absolute shrinkage and selection operator (LASSO) regression to construct a radiomics model. CD38 expression was quantified by immunohistochemistry and combined with radiomics in a fusion model. An independent validation cohort (n=89) was derived from The Cancer Imaging Archive (TCIA) and Gene Expression Omnibus (GEO) datasets. Model performance was evaluated using receiver operating characteristic (ROC) curves, calibration curves, decision curve analysis (DCA), net reclassification improvement (NRI), and integrated discrimination improvement (IDI). Overall survival (OS) differences were assessed by Kaplan-Meier analysis. Molecular mechanisms were explored using differential gene expression, Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment, Cell-type Identification By Estimating Relative Subsets Of RNA Transcripts (CIBERSORT), and single-sample gene set enrichment analysis (ssGSEA).ResultsThe radiomics model achieved area under the curves (AUCs) of 0.734 and 0.878 in the training and validation cohorts, respectively. Incorporating CD38 increased the AUC in the training cohort to 0.801, with significant improvement in NRI (0.952, P<0.001) and IDI (0.148, P=0.001). A nomogram integrating clinical variables enabled individualized prediction. High-response patients had significantly longer OS than low-response patients [hazard ratio (HR) =0.45, P=0.001], with consistent advantages in subgroups such as age >64 years and T1–2 stage. Mechanistically, high responders exhibited enrichment of epithelial differentiation genes and T-cell activation pathways, with increased CD8+ T cells and M1 macrophages; low-response patients were enriched in extracellular matrix (ECM) remodeling genes and collagen metabolism pathways, with higher infiltration of M2 macrophages and regulatory T cells.ConclusionsIntegrating radiomic features with CD38 expression significantly enhances specificity and clinical applicability in predicting immunotherapy response, while elucidating molecular and immunological mechanisms underlying response heterogeneity. The model supports individualized risk stratification and precision therapy planning; however, larger multicenter prospective studies and multi-omics integration are warranted to further validate and optimize its utility.

Personalized repetitive transcranial magnetic stimulation (rTMS) increasingly relies on resting-state functional magnetic resonance imaging (fMRI) to select stimulation sites, yet most pipelines depend on user-defined thresholds and atlas masks, which can shift individualized targets. We propose a watershed-based approach, implemented in a graphical user interface, that performs threshold-independent segmentation of functional images to support rTMS target localization. As a proof-of-concept, we focused on Alzheimer's disease-related circuits within the default mode network, designating the posterior cingulate cortex (PCC) as the deep effective region and the inferior parietal lobule (IPL) as the superficial stimulation target. In a cohort of 21 healthy participants, quantitative comparison with a conventional threshold-based, mask-constrained peak strategy revealed high concordance for PCC peaks but a median spatial displacement of 6.0 mm (95 % CI: 0.0-12.7 mm) for IPL targets. Qualitative examples further illustrate that watershed segmentation reduces bias from neighboring functional clusters, truncation by atlas boundaries, and ambiguity among multiple local peaks. By decoupling target definition from user-chosen thresholds and packaging the method in an accessible toolbox, this framework offers a generalizable tool for individualized fMRI-guided rTMS.

Ferroptosis, an iron-dependent, lipid peroxidation-driven programmed cell death, holds substantial promise for cancer therapy, yet its translational potential is hindered by widespread intrinsic resistance. While glutathione peroxidase 4 (GPX4) is a well-established ferroptosis suppressor, the epigenetic circuitry coordinating GPX4-related mechanisms remains elusive. Here, via genome-wide screening, we identify ten-eleven translocation 1 (TET1)-a key mediator of DNA 5-hydroxymethylation-as a master controller of cancer cell ferroptosis susceptibility. In acute myeloid leukemia (AML), TET1 enhances 5hmC deposition at the glutamate-cysteine ligase catalytic subunit (GCLC) promoter to activate glutathione/γ-glutamyl-peptide metabolism, fortifying GPX4-dependent defense. Concurrently, TET1 activates NFκB signaling to upregulate GTP cyclohydrolase-1 (GCH1), conferring GPX4-independent ferroptosis resistance. Critically, co-targeting TET1/GCLC/GCH1 with low-dose ferroptosis inducers exhibits potent therapeutic effects against both ferroptosis-sensitive and -resistant AML. Our work positions TET1 as a pivotal epigenetic hub governing ferroptosis surveillance, and provides a translatable strategy to overcome ferroptosis resistance in cancer, with AML as a paradigm.