ABSTRACT The Extracellular Vesicles Gene‐based Prostate Score (EGPS), powered by DeepSeek, is an artificial intelligence (AI) diagnostic tool that enhances the detection of clinically significant prostate cancer (csPCa) using urinary EV‐derived gene expression, without requiring digital rectal examination (DRE). To address overdiagnosis resulting from the limited specificity of prostate‐specific antigen (PSA) and reduce unnecessary biopsies, this study evaluated the clinical utility and generalizability of EGPS in men undergoing initial biopsy with PSA levels ranging from 0 to 15 ng/mL. A total of 645 patients were retrospectively enrolled: 586 from three centres were divided into training (70%) and internal validation (30%) cohorts, and 59 from two centres served as the external validation cohort. EVs were isolated using the EXODUS platform, and gene expression was measured by RT‐qPCR. Ten machine learning algorithms were evaluated for constructing the EGPS model with selected genes. Diagnostic efficacy was assessed by ROC analysis, DeLong tests, and decision curve analysis. An AI diagnostic system using DeepSeek was also developed. The EGPS model, incorporating AMACR, HOXB13, and PSGR, achieved AUCs of 0.838, 0.825, and 0.811 in the training, internal validation, and external validation cohorts, respectively, outperforming PSA. At a cut‐off value of 0.22, the model demonstrated sensitivity above 95%, with a missed diagnosis rate of 3.81% in the training cohort and 0% in the validation cohorts. The model reduced unnecessary biopsies by 79 (23.37%), 27 (18.62%) and 9 (15.25%) cases across the three cohorts, thereby lowering biopsy‐related risks. A DeepSeek‐powered AI diagnostic system integrating EGPS was developed to support csPCa diagnosis and minimize unnecessary biopsies. EGPS, derived from multicentre Chinese cohorts, enables accurate, DRE‐free, non‐invasive prediction of csPCa in men with PSA levels of 0–15 ng/mL. When integrated into an AI system, EGPS supports early screening and personalized clinical decision‐making by reducing unnecessary biopsies.
The global rise in steatotic liver disease poses a significant public health challenge. While non-contrast computed tomography scans hold promise for opportunistic detection of steatotic liver disease, their potential for staging and risk assessment remains underexplored. Here we present a multimodal AI model trained on a large dataset, comprising of (n=968) histopathologically and (n=1103) radiologically confirmed cases, validated against both histology (n=660) and MRI-PDFF (n=375) gold standards, demonstrating high accuracy in detecting mild to severe steatosis (AUC: 0.904-0.929) and clinically significant fibrosis (AUC: 0.824-0.888). Furthermore, integrating the model into the standard clinical pathway improves primary risk screening in a retrospective patient cohort (n=1192), identifying 36% more patients at risk of fibrosis progression. Using Cox proportional hazard model, we observe that the intermediate-high risk patients identified by the optimized clinical pathway exhibits a significantly higher incidence of cirrhosis (hazard ratio: 5.54: 2.69-11.42), showcasing the model's potential for early detection and management of steatotic liver disease.
Middle meningeal artery embolization (MMAE) is a promising new treatment for chronic SDH (cSDH). We conducted an international multi-disciplinary survey among physicians involved in the care of cSDH patients to understand the perceived safety profile and efficacy of MMAE in real-world clinical practice. The survey was distributed via several neurointerventional society mailing lists. We collected data on participant demographics, practice patterns, volumes and experience, perceived MMAE safety, complication tolerance, and decision-making. Data were analyzed using descriptive statistics, non-parametric tests and multi-variable regression. Five-hundred-seventy-seven physicians completed the survey between November 20th, 2024 and May 1st, 2025 (median age 42.0 years, 489 [84.7
ZDHHC4, a palmitoyl transferase belonging to the DHHC family, is crucial in cells by catalyzing palmitoylation of proteins, thereby regulating their function and localization. However, its role in melanoma is not well understood. Here, our research determined that ZDHHC4 expression was upregulation in human melanoma specimens and cells. Functional studies reveal that knocking down ZDHHC4 inhibits cell proliferation, migration and invasion of melanoma cells. Moreover, we performed mass spectrometry analysis and found that ZEB-2 is a substrate of ZDHHC4. ZEB-2 interacted with ZDHHC4 through its N-terminal sequences, which promotes the ZDHHC4-mediated palmitoylation of ZEB-2 at C478, facilitating ZEB-2 deubiquitination and its protein stability. This key modification is required for epithelial-to-mesenchymal transition (EMT) in melanoma cells. Furthermore, we found a positive correlation between the expression levels of ZDHHC4 and ZEB-2 in clinical melanoma samples. In summary, our results provide a deeper understanding of mechanism regulating ZDHHC4 in melanoma, suggesting that targeting ZDHHC4 may offer novel therapeutic strategies by suppressing tumor growth and metastasis through the disruption of ZEB-2 palmitoylation process.
AIMS:The burden of chronic liver disease (CLD) is increasing. This study aims to identify protein markers for CLD and its progression, and develop a protein-based risk prediction model. MATERIALS AND METHODS:We used proteome-wide Mendelian randomization (MR), Bayesian colocalization and summary-data-based MR with proteomic data from deCODE Genetics to identify CLD-related proteins. Multivariable MR was used to assess independent protein effects. Protein-protein interaction, druggability and mediation analyses were conducted to prioritise therapeutic targets. We further constructed a protein risk score to predict CLD and composite hepatic event outcomes, comparing its performance with existing clinical predictors. RESULTS:Genetically predicted levels of 16, 5 and 4 plasma proteins were associated with metabolic dysfunction-associated steatotic liver disease (MASLD), alcoholic liver disease (ALD) and cirrhosis, respectively. IGSF3, FTCD, DCXR, ADH1B and ACY1 were associated with CLD progression. Genetically predicted five modifiable factors (body mass index, waist-hip ratio, glycated haemoglobin, type 2 diabetes, leisure television watching) were associated with CLD-related proteins. Proteomic-based models showed high predictive performance for ALD (C-index = 0.89), liver cancer (C-index = 0.84) and liver failure (C-index = 0.84) in a healthy population without liver diseases at baseline. CONCLUSIONS:This study identified key circulating protein markers for CLD. Protein-based profiling demonstrated strong predictive potential for CLD and related outcomes.
