Guanqiu Chen

Objective: To evaluate the triage accuracy, response quality, safety, and repeatability of widely used large language models (LLMs) in urologic emergencies and to examine their potential clinical utility and risks. Methods: We developed 50 standardized clinical vignettes covering 10 categories of urologic emergencies, including testicular torsion, acute urinary retention, and Fournier gangrene. The set comprised high-risk vignettes requiring immediate emergency care, borderline vignettes, and atypical vignettes containing misleading cues. Six commercially available LLMs—Kimi 3.0, DeepSeek V4, Qwen 3.8-MAX, GPT-5.6 Sol (ChatGPT), Claude Opus 4.8, and Ant Afu (AQ)—were evaluated in Chinese-language interactions. Each vignette was submitted three times in independent conversations. Two urologists with more than 10 years of clinical experience assessed triage accuracy using a three-level reference standard (immediate emergency department [ED] evaluation, routine outpatient evaluation, or home observation). They also rated response quality across eight dimensions using a 5-point Likert scale and evaluated medical-error severity (0–3), response stability, and Chinese readability. Overall differences were assessed using the Friedman test, followed by pairwise Wilcoxon signed-rank tests with Holm adjustment. Results: A total of 900 responses were generated. Overall three-class accuracy was 89.11%, and the macro-averaged F1 score was 85.73%. Qwen achieved the highest accuracy (94.00%), whereas Claude achieved the lowest (80.67%). However, none of the pairwise comparisons between models reached statistical significance after Holm adjustment. All models generally recognized scenarios requiring immediate ED evaluation, but their ability to distinguish routine outpatient evaluation from home observation varied substantially. Mean content-quality scores ranged from 3.68 to 4.15; AQ achieved the highest overall score (4.15 ± 0.49), whereas DeepSeek achieved the lowest (3.68 ± 0.48). Clinically relevant errors occurred in 12.3% of responses, including major errors in 6.2% and dangerous errors in 3.1%. Three-run exact triage agreement ranged from 78% to 96%. Readability performance varied by metric, with no single model consistently outperforming the others. Conclusions: The six LLMs demonstrated some ability to perform three-level triage of urologic emergencies, supporting further investigation of their potential clinical applications. Nevertheless, important limitations remained in the management of borderline cases, response stability, and safety; some responses included potentially dangerous self-management instructions or could delay care for high-risk patients. Integration with authoritative guidelines, structured risk warnings, and human review will be necessary to improve clinical safety and reliability.

Medical adhesives have been extensively applied for skin wound repair and hemostasis, while the dynamically humid and complex intraperitoneal environment imposes stringent requirements on tissue adhesives. Few studies have reported the application of biological adhesives for bladder defect repair in urological practice. The GSF composite hydrogel adhesive, composed of gelatin methacryloyl (GelMA), silk fibroin methacryloyl (SilMA), and polyether F127 diacrylate (F127DA), was previously fabricated with favorable injectability, appropriate viscosity and superior mechanical toughness. To further confirm the repair efficacy, GSF was characterized in vitro for the adhesive properties and structural stability under dynamically humid conditions. Corresponding in vivo experiments were conducted on rat models with graded bladder rupture and partial bladder defects to further assess the repair performance of GSF. Results showed that GSF hydrogel retained robust interfacial adhesion and intact structural stability for 8 days under dynamically simulated humid intraperitoneal conditions and cyclic urine filling-emptying environments. GSF exhibited rapid bonding capacity and effectively shortened the operation time. Under moderate bladder injury conditions, the wound closure time of the GSF group was only 28.3% of that of the suture group in cystotomy experiment and 34.5% in cystectomy experiment. It achieved satisfactory repair of mild-to-moderate bladder ruptures and partial bladder defects. Postoperative histological and functional evaluations confirmed that GSF hydrogel facilitated bladder structural reconstruction and functional recovery. These findings highlight the significant potential of GSF hydrogel adhesives for repairing and regenerating injured bladders. Future applications of GSF will extend to adhesive repairs of organs within the humidity conditions of the abdominal cavity.

Background: Necroptosis has been reported to play a critical role in occurrence and progression of cancer. The dysregulation of long non-coding RNAs (lncRNAs) is associated with the progression and metastasis of clear cell renal cell carcinoma (CCRCC). However, research on necroptosis-related lncRNAs in the tumor heterogeneity and prognosis of CCRCC is not completely unclear. This study aimed to analysis the tumor heterogeneity among CCRCC subgroups and construct a CCRCC prognostic signature based on necroptosis-related lncRNAs. Methods: Weighted gene co-expression network analysis (WGCNA) was performed to identify necroptosis-related lncRNAs. A preliminary classification of molecular subgroups was performed by non-negative matrix factorization (NMF) consensus clustering analysis. Comprehensive analyses, including fraction genome altered (FGA), tumor mutational burden (TMB), DNA methylation alterations, copy number variations (CNVs), and single nucleotide polymorphisms (SNPs), were performed to explore the potential factors for tumor heterogeneity among the three subgroups. Subsequently, we constructed a predictive signature by multivariate Cox regression. Nomogram, calibration curves, decision curve analysis (DCA), and time-dependent receiver-operating characteristics (ROC) were used to validate and evaluate the signature. Finally, immune correlation analyses, including immune-related signaling pathways, immune cell infiltration status and immune checkpoint gene expression level, were also performed. Results: Seven necroptosis-related lncRNAs were screened out by WGCNA, and three subgroups were classified by NMF consensus clustering analysis. There were significant differences in survival prognosis, clinicopathological characteristics, enrichments of immune-related signaling pathway, degree of immune cell infiltration, and expression of immune checkpoint genes in the various subgroups. Most importantly, we found that 26 differentially expressed genes (DEGs) among the 3 subgroups were not affected by DNA methylation alterations, CNVs and SNPs. On the contrary, these DEGs were associated with the seven necroptosis-related lncRNAs. Subsequently, the identified RP11-133F8.2 and RP11-283G6.4 by multivariate Cox regression analysis were involved in the risk model, which could serve as an independent prognostic factor for CCRCC. Finally, qRT-PCR confirmed the differential expression of the two lncRNAs. Conclusions: These findings contributed to understanding the function of necroptosis-related lncRNAs in CCRCC and provided new insights of prognostic evaluation and optimal therapeutic strategy for CCRCC.