With the rapid advancement of artificial intelligence (AI) technology, particularly breakthroughs in large models, AI applications in healthcare are becoming increasingly widespread.Ophthalmology, as a critical branch of medicine, has emerged as a clinical specialty with comprehensive AI research and applications, especially in the AI-driven analysis of ophthalmic imaging.Currently, AI in ophthalmology is undergoing profound transformation driven by large model technology.AI large models, with their robust data comprehension and multimodal interaction capabilities, offer new possibilities for the diagnosis, treatment, and health education of ophthalmic diseases, powerfully propelling the evolution of current AI paradigms in ophthalmology.This article explores the novel opportunities brought by the extensive application of AI large models in ophthalmic research and practice, and provides recommendations for ophthalmologists and insights for industry development.

To investigate the clinical characteristics,genetic mutation patterns,and therapeutic strategies of an infant with interstitial lung disease associated with a surfactant protein C(SFTPC)gene mutation,and systematic review of 188 cases of SFTPC mutation-elated childhood ILD(chILD)reported in Chinese and English literature(from database inception to December 2024)was conducted.The patient in this study was a girl aged 11 months.She presented with progressive dyspnea,hypoxemia,and severe growth retardation starting at 2 months of age.Whole-exome sequencing identified a heterozygous missense mutation in SFTPC(c.187A>G,p.K63E),inherited from her mother.Symptoms significantly improved after treatment with glucocorticoids and azithromycin,though pulmonary bullae persisted at 5-year follow-up.Literature analysis revealed that the hotspot mutation I73T was located in the linker domain,while BRICHOS domain mutations exhibited the highest genetic diversity.BRICHOS domain mutations were associated with earlier onset,whereas non-BRICHOS mutations predominantly manifested in infancy with chronic cough,hypoxemia,and growth retardation.Oxygen therapy and glucocorticoids were the most common treatments.Hydroxychloroquine was more frequently used in linker domain mutations.The overall survival rate was 84%.SFTPC mutations are a critical genetic etiology of chILD in infants.Early genetic testing combined with glucocorticoid/macrolide therapy improves prognosis,but long-term monitoring for structural lung damage(e.g.,pulmonary bullae)is essential.

With the rapid advancement of artificial intelligence (AI) technology, particularly breakthroughs in large models, AI applications in healthcare are becoming increasingly widespread.Ophthalmology, as a critical branch of medicine, has emerged as a clinical specialty with comprehensive AI research and applications, especially in the AI-driven analysis of ophthalmic imaging.Currently, AI in ophthalmology is undergoing profound transformation driven by large model technology.AI large models, with their robust data comprehension and multimodal interaction capabilities, offer new possibilities for the diagnosis, treatment, and health education of ophthalmic diseases, powerfully propelling the evolution of current AI paradigms in ophthalmology.This article explores the novel opportunities brought by the extensive application of AI large models in ophthalmic research and practice, and provides recommendations for ophthalmologists and insights for industry development.

To investigate the clinical characteristics,genetic mutation patterns,and therapeutic strategies of an infant with interstitial lung disease associated with a surfactant protein C(SFTPC)gene mutation,and systematic review of 188 cases of SFTPC mutation-elated childhood ILD(chILD)reported in Chinese and English literature(from database inception to December 2024)was conducted.The patient in this study was a girl aged 11 months.She presented with progressive dyspnea,hypoxemia,and severe growth retardation starting at 2 months of age.Whole-exome sequencing identified a heterozygous missense mutation in SFTPC(c.187A>G,p.K63E),inherited from her mother.Symptoms significantly improved after treatment with glucocorticoids and azithromycin,though pulmonary bullae persisted at 5-year follow-up.Literature analysis revealed that the hotspot mutation I73T was located in the linker domain,while BRICHOS domain mutations exhibited the highest genetic diversity.BRICHOS domain mutations were associated with earlier onset,whereas non-BRICHOS mutations predominantly manifested in infancy with chronic cough,hypoxemia,and growth retardation.Oxygen therapy and glucocorticoids were the most common treatments.Hydroxychloroquine was more frequently used in linker domain mutations.The overall survival rate was 84%.SFTPC mutations are a critical genetic etiology of chILD in infants.Early genetic testing combined with glucocorticoid/macrolide therapy improves prognosis,but long-term monitoring for structural lung damage(e.g.,pulmonary bullae)is essential.

Objective To explore the effectiveness of high-definition diffusion weighted imaging(DWI)sequence combined with T1 WI-fat suppression(FS)dynamic contrast enhancement(DCE)sequence for preoperative T-stage of rectal cancer by using 3.0T MRI standardized scanning.Methods A retrospective analysis was conducted on MRI images of 57 patients with rectal cancer confirmed by pathology.Before surgery,the patients underwent 3.0T MRI standardized rectal cancer scan methods,including routine sequence,high-definition DWI sequence,and T1 WI-FS DCE sequence,etc.Then two experienced physicians evaluated the T-stage of preoperative rectal cancer through high-definition DWI(transverse and sagittal sections)and T1 WI-FS DCE sequences in the double-blind method.Using the postoperative pathological results of rectal cancer as the"gold standard",two sequences were combined to evaluate the accuracy,sensitivity,and specificity of rectal cancer T-stage.Results Among the 57 cases,there were 9 cases of upper rectal cancer,39 cases of middle rectal cancer,and 9 cases of lower rectal cancer.The accuracy rates of preoperative T-stage diagnosis for rectal cancer by two evaluator were both 85.7％(6/7)in T1 stage,88.2％(15/17)and 94.1％(16/17)in T2 stage,96.9％(31/32)and 93.8％(30/32)in T3 stage,and both 100.0％(1/1)in T4 stage.For evaluator 1,the sensitivity and specificity of the rectal cancer T-stage diagnosis were 96.1％and 83.3％,and for evaluator 2 were 94.1％and 83.3％,respectively.For rectal cancer MRI diagnosis,the accuracy rates and sensitivity were higher when combining the high-definition DWI sequence and T1 WI-FS DCE sequence,compared with a single high-definition DWI sequence or T1 WI-FS DCE sequence,and the difference was statistically significant.The average preoperative apparent diffusion coefficient(ADC)value of rectal cancer was compared between the corresponding postoperative pathological T1 to T4 stage groups,and the difference was statistically significant.Conclusion The combination of high-definition DWI sequence and T1 WI-FS DCE sequence improves the accuracy of rectal cancer T-stage,providing assistance for personalized clinical treatment.

With the accelerated aging society in China, the incidence of biliary surgical diseases in the elderly has increased significantly. The clinical characteristics of these patients indicate that improving treatment outcomes and realizing healthy aging are worthy of attention. How to effectively improve the treatment effect of geriatric biliary surgical diseases has attracted widespread attention. This paper reviews and comments on the hotspots and difficulties of biliary surgery in older patients from six aspects: (1) higher morbidity associated with an aging society, (2) prevention and control of pre-operative risks, (3) extending the indications of laparoscopic surgery, (4) urgent standardization of minimally invasive surgery, (5) precise technological progress in hepatobiliary surgery, and (6) guarantee of peri-operative safety. It is of great significance to fully understand the focus of controversy, actively make use of its favorable factors, and effectively avoid its unfavorable factors, for further improving the therapeutic effects of geriatric biliary surgical diseases, and thus benefits the vast older patients with biliary surgical diseases. Accordingly, a historical record with the highest age of 93 years for laparoscopic transcystic common bile duct exploration has been created by us recently.
Humans ; Aged ; Aged, 80 and over ; Biliary Tract Surgical Procedures ; Gallstones ; Laparoscopy ; Treatment Outcome ; Aging ; Retrospective Studies

Objective:To analyze the change of differential genes and signaling pathways in high glucose induced BV2 cells, and to explore the mechanism of transgelin-2 (TAGLN2) regulating cellular inflammatory response and metabolic process.Methods:An experimental study. The cultured BV2 cells were divided into mannitol treatment (Man) group, glucose treatment (Glu) group, overexpression control Glu treatment (Con) group, overexpression TAGLN2 Glu treatment group, silence control Glu treatment (shCon Glu) group, and silence TAGLN2 Glu treatment (shTAGLN2 Glu) group. Cells in the Man group were cultured in modified Eagle high glucose medium (DMEM) containing 25 mmol/L mannitol and 25 mmol/L glucose, cells in other groups (Glu group, Con Glu group, TAGLN2 Glu group, shCon Glu group and shTAGLN2 Glu group) were cultured in DMEM medium containing 50 mmol/L glucose. After 24 hours of cells culture, transcriptome sequencing of cells in each group were performed using high-throughput sequencing technology, and significantly differentially expressed genes (DEG) were screened. |log 2 (fold change)|≥1 and P≤0.05 were adopted as criteria to screen for DEG. Gene Ontology (GO) function enrichment analysis, Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis and protein-protein interaction network analysis were performed. Real-time polymerase chain reaction (RT-PCR) was used to detect the relative expression level of DEG mRNA. The data between groups were compared by independent sample t-test. Results:When compared with Man group, a total of 517 differentially expressed genes were screened in Glu group, which including 277 up-regulated genes and 240 down-regulated genes. KEGG pathway enrichment analysis showed that the up-regulated genes were significantly enriched in immune system processes such as nuclear factor (NF)-κB signal pathway, Jak-signal transducers and activators of transcription (STAT) signal pathway, while down-regulated genes were significantly enriched in glycosaminoglycan degradation and glyceride metabolic pathway. Compared with Con Glu group, a total of 480 DEG were screened in TAGLN2 Glu group, among which 147 up-regulated and 333 down-regulated genes were detected. Up-regulated genes were significantly enriched in the metabolic processes of fatty acid, glyceride and pyruvate, while down-regulated genes were significantly enriched in immune system processes such as NF-κB signal pathway, Jak-STAT signal pathway and tumor necrosis factor (TNF) signal pathway. Compared with shCon Glu group, a total of 582 DEG were screened in shTAGLN2 Glu group, among which 423 up-regulated and 159 down-regulated genes were detected. Up-regulated DEG were significantly enriched in immune system processes such as TNF signal pathway and chemokine signal pathway, while down-regulated DEG were significantly enriched in pattern recognition receptor signal pathway. RT-PCR results showed that the relative expression levels of DEG mRNA Card11 ( t=13.530), Icos ( t=3.482), Chst3 ( t=6.949), Kynu ( t=5.399), interleukin (IL)-1β ( t=2.960), TNF-α ( t=5.800), IL-6 ( t=3.130), interferon-γ ( t=7.690) and IL-17 ( t=6.530) in the TAGLN2 Glu treatment group were decreased significantly compared with Con Glu group, and the difference was statistically significant. Conclusion:TAGLN2 can inhibit glucose induced microglia inflammation by NF-κB and Jak-STAT signaling pathways, Card11, Icos, Chst3 and Kynu play an important role in the anti-inflammatory process of TAGLN2.

Objective:To analyze the association between high density lipoprotein cholesterol (HDL-C) and the risk of cardiovascular disease mortality.Methods:A total of 71 618 residents aged over 18 years with complete baseline data, who were filed on the health information big data platform of Yinzhou district, Ningbo city, Zhejiang Province from 2009 to 2014, were selected as the research population. The research population were divided into four groups according to the level of HDL-C: low-level group (HDL-C<1.0 mmol/L), intermediate-level group (1.0 mmol/L≤HDL-C<1.5 mmol/L), medium-high-level group (1.5 mmol/L≤HDL-C<2.0 mmol/L) and high-level group (HDL-C≥2.0 mmol/L). Cox proportional hazard model was used to calculate the risk ratio of cardiovascular diseases mortality in different groups.Results:The study population was followed up for a total of 427 989.4 person-years, follow-up time of (5.98±1.04)years. During the follow-up period, there were 799 deaths due to cardiovascular diseases. After adjusting for confounding factors, compared with the medium-high-level group as the reference group, the HR (95% CI) for cardiovascular diseases mortality was 1.43 (1.13-1.82) in the low-level group and 1.22 (1.02-1.46) in the high-level group. Conclusion:The low level of HDL-C (<1.5 mmol/L) is associated with a higher risk of cardiovascular disease deaths. The level of HDL-C can be used as a biological indicator to monitor the development of cardiovascular diseases and guide treatment.

Objective:To analyze the association between high density lipoprotein cholesterol (HDL-C) and the risk of cardiovascular disease mortality.Methods:A total of 71 618 residents aged over 18 years with complete baseline data, who were filed on the health information big data platform of Yinzhou district, Ningbo city, Zhejiang Province from 2009 to 2014, were selected as the research population. The research population were divided into four groups according to the level of HDL-C: low-level group (HDL-C<1.0 mmol/L), intermediate-level group (1.0 mmol/L≤HDL-C<1.5 mmol/L), medium-high-level group (1.5 mmol/L≤HDL-C<2.0 mmol/L) and high-level group (HDL-C≥2.0 mmol/L). Cox proportional hazard model was used to calculate the risk ratio of cardiovascular diseases mortality in different groups.Results:The study population was followed up for a total of 427 989.4 person-years, follow-up time of (5.98±1.04)years. During the follow-up period, there were 799 deaths due to cardiovascular diseases. After adjusting for confounding factors, compared with the medium-high-level group as the reference group, the HR (95% CI) for cardiovascular diseases mortality was 1.43 (1.13-1.82) in the low-level group and 1.22 (1.02-1.46) in the high-level group. Conclusion:The low level of HDL-C (<1.5 mmol/L) is associated with a higher risk of cardiovascular disease deaths. The level of HDL-C can be used as a biological indicator to monitor the development of cardiovascular diseases and guide treatment.

Objective:To analyze differentially expressed genes (DEGs) and the changes of signal pathways in human retinal pigment epithelium cells (ARPE-19) under hypoxic and normoxic conditions and to explore the biological mechanism of hypoxia-induced ARPE-19 cell damage via transcriptome sequencing (RNA-seq) and bioinformatics technology.Methods:The ARPE-19 cells were divided into the hypoxia treatment group and the normoxia control group treated with 1% and 21% O 2 by volume for 8, 24, 48, 72 hours, respectively.The relative expression levels of vascular endothelial growth factor (VEGF) and hypoxia-inducible factor-1α (HIF-1α) mRNA were detected with real-time fluorescent quantitative PCR at different time points.RNA-seq and bioinformatics analysis were performed at 8 hours and 24 hours after hypoxia and normoxia treatment.DEGs were screened out under the conditions of |log 2FC|≥1 and P≤0.05.Then the cluster heat map analysis, Gene Ontology (GO) functional enrichment analysis, Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis and protein-protein interaction network analysis were also carried out.Real-time fluorescent quantitative PCR was employed at 24 hours after hypoxia to detect the relative mRNA expression of genes that might be related to hypoxia in DEGs.Cell viability kit was used to verify and compare the damage effect of hypoxia on ARPE-19 cells at different time points between the two groups. Results:The relative mRNA expression levels of VEGF at 8, 24, 48 and 72 hours after hypoxia treatment and the relative HIF-1α mRNA expression levels at 8, 24 and 48 hours after hypoxia treatment were significantly higher than those of the normoxia control group (all at P<0.05). There were large differences in the mRNA expression levels at 8-hour and 24-hour treatment between the two groups.A total of 62 significant DEGs were screened between the hypoxia treatment group and the normoxia control group after 8-hour hypoxia treatment, among which 45 genes were significantly up-regulated and 17 genes were significantly down-regulated.A total of 255 significant DEGs were screened out between the hypoxia treatment group and the normoxia control group after 24-hour hypoxia treatment, among which 228 genes were significantly up-regulated and 27 genes were significantly down-regulated.The GO functional analysis of DEGs was mainly enriched in processes such as protein degradation, nucleotide biosynthesis, and material transport.KEGG pathway analysis was mainly enriched in PI3K-Akt, cGMP-PKG, and other signaling pathways closely related to metabolism, cell cycle, cell growth, and apoptosis.The core genes HPCA, MT3 and NOS3 were found by protein-protein interaction network analysis.Real-time fluorescent quantitative PCR test results showed that after 24-hour hypoxia treatment, the mRNA expression levels of hypoxia related genes DEPP1, NPPB, PDZK1, HILPDA, TCEA3, NDRG1 and RORC in ARPE-19 cells were significantly increased and the mRNA expression levels of TFRC and NQO1 were significantly decreased (all at P<0.05). The cell morphology was normal and the growth state was good without dead cells after 8-hour and 24-hour hypoxia treatment in ARPE-19 cells.There were dead cells after 48-hour hypoxia treatment, and the number of dead cells was increased at 72 hours after hypoxia treatment. Conclusions:The PI3K-Akt and cGMP-PKG signaling pathways related to metabolism may be involved in hypoxia-induced injury of ARPE-19 cells.Core genes of HPCA, MT3 and NOS3 can be used as functional target genes and play key roles in hypoxia response of cells.