学术周报 · IF≥10 或专科旗舰刊

眼科领域文献阅读汇编

2026年第46周 (2026-11-13) | PubMed (NLM) · DeepSeek 中英双语
数据来源: PubMed E-utilities · 影响因子筛选: IF≥10 或专科旗舰刊 · 完整摘要不截断
收录论文
14
临床研究
7
基础研究
7
IF≥20
0
IF 10-20
12
子领域
6
期刊种类
10
数据日期
2026-11-13

★本周 Top 10 高影响力文献

#论文期刊IF
1Engineered compact TnpB epigenetic editors prevent choroidal neovascularization via Vegfa repression...Protein & cellIF 18.2
2Acute Zonal Occult Outer Retinopathy: Reappraising a Historical Clinical Concept in the Era of Multi...Progress in retinal and eye researchIF 16.2
3Advances in Retinal Cellular Imaging: Towards Single-Cell Resolution and Functional Biomarkers.Progress in retinal and eye researchIF 16.2
4Standardising ocular surface photography for artificial intelligence: current limitations and a prop...Progress in retinal and eye researchIF 16.2
5Memory-guided mutual information optimization in retinal imaging for early stroke detection and subt...Medical image analysisIF 14.0
6On-going retinal neuronal degeneration despite clinical stability in glaucoma.Science advancesIF 13.9
7The Neurovascular Unit in Diabetic Retinopathy: The Emerging Role of Pyroptosis and Therapeutic Pros...Medicinal research reviewsIF 13.6
8Emerging strategies for progressive supranuclear palsy (PSP) combination therapies.EBioMedicineIF 11.2
9Step change in glaucoma polygenic risk score performance enables clinical utility.OphthalmologyIF 10.9
10Four-Year Outcomes of Faricimab in Diabetic Macular Edema: Results from the RHONE-X Extension Trial.OphthalmologyIF 10.9

Ŧ期刊分布统计

期刊篇数IF
Progress in retinal and eye research3IF 16.2
Ophthalmology2IF 10.9
American journal of ophthalmology2IF 4.7
Protein & cell1IF 18.2
JAMA ophthalmology1IF 10.5
Journal of hazardous materials1IF 10.6
EBioMedicine1IF 11.2
Medical image analysis1IF 14.0
Medicinal research reviews1IF 13.6
Science advances1IF 13.9

1视网膜疾病 (5篇)

临床研究 (2篇)

Ophthalmology IF 10.9 2026-11-10 PMID: 43013386
The RHONE-X study (ClinicalTrials.gov identifier, NCT04432831) evaluated long-term safety and tolerability (primary end points), and efficacy (exploratory end points) of faricimab, a dual angiopoietin-2 (Ang-2) and vascular endothelial growth factor A (VEGF-A) inhibitor, using a treat-and-extend (T&E) protocol in patients with diabetic macular edema (DME). Global phase 3, multicenter, nonrandomized, 2-year open-label extension study of the YOSEMITE and RHINE (ClinicalTrials.gov identifiers, NCT03622580 and NCT03622593, respectively) studies. Of 1622 patients who completed YOSEMITE and RHINE, 1474 patients (91%) were included in the RHONE-X trial. Patients transitioned from faricimab every 8 weeks (Q8W), faricimab T&E, or aflibercept 2.0 mg Q8W in YOSEMITE and RHINE to the RHONE-X study without requiring monthly initiation doses. All received faricimab T&E at up to 16-week intervals based on best-corrected visual acuity (BCVA) and central subfield thickness (CST), per YOSEMITE and RHINE criteria. Patients attended every month during the initial 4-month masked period, then only for T&E dosing visits (open-label arm). Analysis windows were defined for the RHONE-X study years 1, 1.5, and 2 to account for visit asynchronicity. The primary end point was incidence and severity of ocular and nonocular adverse events (AEs). Exploratory end points included change from baseline BCVA and CST, proportion of patients with absence of DME (CST < 325 μm), and treatment durability. Overall, 1204 patients (82%) completed the RHONE-X trial. The incidence of AEs leading to treatment discontinuation (1.5%) and rates of intraocular inflammation (1.3%) were low. Overall, faricimab T&E maintained visual and anatomic improvements achieved in YOSEMITE and RHINE. Adjusted mean BCVA improvements from YOSEMITE and RHINE baseline to the end of the RHONE-X trial were +10.1 letters (faricimab T&E), +11.4 letters (faricimab T&E [prior Q8W]), and +9.5 letters (faricimab T&E [prior aflibercept]); CST reductions were -198.3 μm, -202.5 μm, and -204.9 μm, respectively. At study end, more than 90% of patients achieved DME absence, regardless of prior treatment. Median number of injections over the RHONE-X trial (2 years) were 7 (faricimab T&E), 8 (faricimab T&E [prior Q8W]), and 8 (faricimab T&E [prior aflibercept]). By the RHONE-X trial completion, approximately 80% of patients received faricimab at ≥Q12W intervals. Faricimab, a dual Ang-2 and VEGF-A inhibitor, provided long-term safety, efficacy, and durability in patients with DME treated using a T&E regimen. Sustained visual and anatomic improvements were observed, with extended dosing intervals and reduced treatment burden. Faricimab was well tolerated, with a safety profile consistent with the YOSEMITE and RHINE trials. Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.
中文摘要:RHONE-X研究(ClinicalTrials.gov标识符,NCT04432831)评估了双血管内皮生长因子A和血管生成素-2抑制剂faricimab在糖尿病黄斑水肿患者中的长期安全性和耐受性(主要终点)及疗效(探索性终点),采用治疗并延长方案。这是YOSEMITE和RHINE研究的多中心、非随机、为期两年的开放标签扩展研究,全球3期。完成YOSEMITE和RHINE研究的1622名患者中,有1474名(91%)纳入RHONE-X试验。患者从YOSEMITE和RHINE中的每8周一次faricimab、faricimab治疗并延长或每8周一次2.0mg阿柏西普过渡到RHONE-X研究,无需每月给予初始剂量。所有患者均接受基于最佳矫正视力和中央亚区厚度的faricimab治疗并延长方案,间隔最长可达16周。患者在最初4个月的盲法期间每月就诊,之后仅在治疗并延长给药访视时就诊(开放标签组)。为考虑访视不同步性,定义了RHONE-X研究第1年、第1.5年和第2年的分析窗口。主要终点是眼部和非眼部不良事件的发生率和严重程度。探索性终点包括最佳矫正视力从基线的变化、中央亚区厚度、无糖尿病黄斑水肿患者比例和治疗持久性。总体而言,1204名患者(82%)完成了RHONE-X试验。导致治疗中断的不良事件发生率为1.5%,眼内炎症发生率为1.3%,均较低。总体而言,faricimab治疗并延长方案维持了YOSEMITE和RHINE研究中实现的视力和解剖学改善。从YOSEMITE和RHINE基线到RHONE-X试验结束的最佳矫正视力调整均值改善分别为+10.1个字母(faricimab治疗并延长)、+11.4个字母(此前每8周一次的faricimab治疗并延长)和+9.5个字母(此前阿柏西普的faricimab治疗并延长);中央亚区厚度减少分别为-198.3μm、-202.5μm和-204.9μm。研究结束时,无论既往治疗如何,超过90%的患者实现了无糖尿病黄斑水肿。RHONE-X试验期间(2年)的中位注射次数分别为7次(faricimab治疗并延长)、8次(此前每8周一次的faricimab治疗并延长)和8次(此前阿柏西普的faricimab治疗并延长)。到RHONE-X试验完成时,约80%的患者接受间隔≥12周的faricimab治疗。Faricimab作为一种双血管生成素-2和血管内皮生长因子A抑制剂,在治疗并延长方案下的糖尿病黄斑水肿患者中提供了长期安全性、疗效和持久性。观察到持续的视力和解剖学改善,给药间隔延长且治疗负担减轻。Faricimab耐受性良好,其安全性特征与YOSEMITE和RHINE试验一致。专有或商业披露可在文章末尾的脚注和披露部分找到。
American journal of ophthalmology IF 4.7 2026-11-07 PMID: 42998694
This systematic review and meta-analysis evaluated the association between statin use and the prevalence, incidence, and progression of age-related macular degeneration (AMD). AMD is a major cause of vision loss in older adults, many of whom receive statins for cardiovascular prevention. We searched PubMed, Embase, Scopus, Web of Science, and the Cochrane Library through October 2025 for randomized trials and cohort, case-control, and cross-sectional studies with comparators. We assessed risk of bias with tools appropriate to each design and pooled effect measures separately using random-effects models. The protocol was registered in PROSPERO. Twenty-eight studies included 1,918,120 participants. Pooled odds ratios (ORs) for the prevalence of any, early, and late AMD were 1.03 (95% confidence interval [CI], 0.86-1.24), 1.10 (95% CI, 0.92-1.31), and 0.75 (95% CI, 0.46-1.25), respectively. For incidence of any AMD, the pooled hazard ratio (HR) was 1.04 (95% CI, 0.82-1.31) and the OR was 0.76 (95% CI, 0.43-1.34), with substantial heterogeneity. For neovascular AMD incidence, different study subsets suggested lower hazard (HR, 0.72; 95% CI, 0.60-0.85) and higher risk (risk ratio, 1.31; 95% CI, 1.18-1.44); the OR was inconclusive (0.93; 95% CI, 0.82-1.05). Progression evidence was limited and heterogeneous. Ten of 12 longitudinal nonrandomized studies had serious overall risk of bias. The evidence was of very low certainty and insufficient to establish a protective or harmful effect of statins on AMD. These findings do not support prescribing statins specifically for AMD or discontinuing otherwise indicated therapy.
中文摘要:本系统综述与荟萃分析评估了他汀类药物使用与年龄相关性黄斑变性(AMD)患病率、发病率及进展之间的关联。AMD是老年人失明的主要原因,许多患者因心血管预防而服用他汀类药物。我们检索了截至2025年10月的PubMed、Embase、Scopus、Web of Science和Cochrane Library数据库,纳入随机试验以及具有对照组的队列研究、病例对照研究和横断面研究。我们采用适用于各设计类型的工具评估偏倚风险,并使用随机效应模型分别合并效应指标。方案已在PROSPERO注册。共纳入28项研究,涉及1918120名参与者。任何AMD、早期AMD和晚期AMD的汇总比值比(OR)分别为1.03(95%置信区间[CI],0.86-1.24)、1.10(95% CI,0.92-1.31)和0.75(95% CI,0.46-1.25)。对于任何AMD的发病率,汇总风险比(HR)为1.04(95% CI,0.82-1.31),OR为0.76(95% CI,0.43-1.34),存在显著的异质性。对于新生血管性AMD的发病率,不同的研究亚组提示风险降低(HR,0.72;95% CI,0.60-0.85)和风险升高(风险比,1.31;95% CI,1.18-1.44);OR结果不明确(0.93;95% CI,0.82-1.05)。进展证据有限且异质性大。12项纵向非随机研究中,有10项存在严重的整体偏倚风险。证据确定性极低,不足以确立他汀类药物对AMD具有保护或有害作用。这些发现不支持为治疗AMD而专门处方他汀类药物,也不支持停用其他指征所需的他汀类药物治疗。

基础研究 (3篇)

Protein & cell IF 18.2 2026-11-13 PMID: 43021643
Epigenetic editing enables programmable and heritable gene regulation without altering DNA sequence, yet existing platforms remain constrained by large size, cumbersome target design, and limited applicability. Here we engineered catalytically inactive ISDge10 and ISAam1 TnpB systems and developed highly compact RNA-guided epigenetic editors compatible with single adeno-associated virus (AAV) delivery for either transcriptional activation or repression. Activators based on these TnpBs (Dge10on and Aam1on) efficiently activated diverse endogenous genes in vitro, and AAV9-delivered Aam1on achieved robust pancreatic and duodenal homeobox 1 (Pdx1) activation in vivo. For gene repression, by screening 248 human Krüppel-associated box (KRAB) domains with verified effector activity, we identified the most potent effector domain for each TnpB system, resulting in the optimization of Dge10off and Aam1off. These epigenetic editors mediated durable transcriptional repression across multiple loci in diverse cell types. In a mouse model, single-AAV delivery of these editors efficiently suppressed vascular endothelial growth factor A (Vegfa) expression with no detectable activation of DNA damage response markers (γH2AX and p53), resulting in marked reductions in choroidal neovascularization area. Our study expands the epigenetic editing toolbox and provides a potentially safe and durable single-administration therapeutic strategy for wet age-related macular degeneration (wAMD).
中文摘要:表观遗传编辑可实现可编程且可遗传的基因调控而不改变DNA序列,但现有平台仍受限于体积庞大、靶点设计繁琐及适用性有限。本研究工程化改造了催化失活的ISDge10和ISAam1 TnpB系统,开发了高度紧凑的RNA引导表观遗传编辑器,兼容单腺相关病毒(AAV)递送,用于转录激活或抑制。基于这些TnpB的激活剂(Dge10on和Aam1on)在体外高效激活了多种内源基因,且AAV9递送的Aam1on在体内实现了胰腺十二指肠同源盒1(Pdx1)的强效激活。对于基因抑制,通过筛选248个人类Krüppel相关盒(KRAB)结构域并验证其效应活性,我们确定了每种TnpB系统最有效的效应结构域,从而优化了Dge10off和Aam1off。这些表观遗传编辑器在多种细胞类型的多个位点介导了持久的转录抑制。在小鼠模型中,单次AAV递送这些编辑器有效抑制了血管内皮生长因子A(Vegfa)的表达,且未检测到DNA损伤反应标志物(γH2AX和p53)的激活,导致脉络膜新生血管面积显著减少。本研究扩展了表观遗传编辑工具库,并为湿性年龄相关性黄斑变性(wAMD)提供了一种潜在安全且持久的单次给药治疗策略。
Journal of hazardous materials IF 10.6 2026-11-10 PMID: 43019986
The increasing coexistence of arsenic (As) and microplastics (MPs) in acid mine drainage (AMD) environments poses amplified ecological risks to aquatic ecosystems. However, the mechanisms underlying their co-immobilization remain poorly understood. Here, quartz crystal microbalance with dissipation monitoring, molecular dynamics simulations, and microanalytical techniques were integrated to investigate Fe(II) organo-mineralization mediated by Acidithiobacillus ferrooxidans, a typical iron-oxidizing bacterium, and its role in As-MPs co-immobilization. The results show that A. ferrooxidans promoted organo-poorly crystalline Fe-As minerals (organo-PCFAMs) formation, whose organic components including lipids, free fatty acids, and/or saccharides, facilitated their adhesion to polyethylene (PE)- and polystyrene (PS)-MPs primarily through van der Waals interactions. Meanwhile, hydroxyl radicals generated during Fe(II) bio-oxidation, continued organo-PCFAMs accumulation and transformation intensified MPs surface damage with their size reduction, and increased organo-PCFAMs loading capacity. These coupled processes enhanced As-MPs co-immobilization, achieving removal rates of 61-98% for As and 73-97% for MPs, with PS-MPs immobilized more effectively than PE-MPs owing to stronger interfacial interactions. Overall, this study elucidates molecular-to-microscale mechanisms by which A. ferrooxidans-mediated organo-mineralization generated mineral bridge to drive As-MPs co-immobilization, advances understanding of their coupled biogeochemical cycling, and provides a strategy for mitigating As-MPs co-contamination in AMD environments.
中文摘要:酸性矿山排水环境中砷与微塑料共存加剧了水生生态系统的生态风险,但其共固定机制尚不明确。本研究整合石英晶体微天平耗散监测、分子动力学模拟及显微分析技术,探究典型铁氧化菌氧化亚铁硫杆菌介导的Fe(II)有机矿物化过程及其在砷与微塑料共固定中的作用。结果显示,该菌促进有机质贫乏的非晶态铁-砷矿物形成,其脂质、游离脂肪酸和/或糖类等有机成分主要通过范德华力促进矿物粘附于聚乙烯和聚苯乙烯微塑料表面。同时,Fe(II)生物氧化产生的羟基自由基持续推动有机非晶态铁-砷矿物的积累与转化,加剧微塑料表面损伤并减小粒径,进而提高有机非晶态铁-砷矿物的负载能力。这些耦合过程增强了砷与微塑料的共固定效果,实现61%至98%的砷去除率和73%至97%的微塑料去除率,其中聚苯乙烯微塑料因更强的界面相互作用而固定效果更佳。本研究阐明了氧化亚铁硫杆菌介导的有机矿物化通过生成矿物桥驱动砷与微塑料共固定的分子至微观尺度机制,深化了对耦合生物地球化学循环的理解,并为缓解酸性矿山排水环境中的砷与微塑料复合污染提供了策略。
Medicinal research reviews IF 13.6 2026-11-09 PMID: 43004652
Diabetic retinopathy (DR) is recognized as a neurovascular disorder, characterized by both functional impairment and structural disruption of the neurovascular unit (NVU), the functional unit of the retina. This review concentrates on the emerging role of pyroptosis, a lytic form of programmed cell death (PCD), within the NVU. We systematically discuss how pyroptosis manifests in the major cell types constituting the NVU, including neurons, Müller cells, microglia, astrocytes, endothelial cells, and pericytes, and delineate the underlying molecular pathways. Furthermore, we integrate recent and emerging evidence associated with pyroptosis and the broader concept of PANoptosis, emphasizing that future diagnostic and therapeutic strategies should transcend a single-pathway perspective to provide optimal protection for the NVU. By integrating recent advances in molecular and cellular research, this review aims to provide novel insights into preserving NVU integrity and developing targeted interventions to delay or halt the progression of DR.
中文摘要:糖尿病视网膜病变(DR)被认定为一种神经血管疾病,其特征是视网膜功能单位——神经血管单元(NVU)的功能障碍和结构破坏。本综述聚焦于作为一种裂解性程序性细胞死亡(PCD)形式的细胞焦亡在 NVU 中的新兴作用。我们系统讨论了细胞焦亡在构成 NVU 的主要细胞类型(包括神经元、Müller 细胞、小胶质细胞、星形胶质细胞、内皮细胞和周细胞)中的表现,并阐明了潜在的分子通路。此外,我们整合了与细胞焦亡及更广泛 PANoptosis 概念相关的最新证据,强调未来的诊断和治疗策略应超越单一路径视角,以提供对 NVU 的最佳保护。通过整合分子和细胞研究的最新进展,本综述旨在为保持 NVU 完整性和开发靶向干预措施以延缓或阻止 DR 进展提供新见解。

2影像与人工智能 (4篇)

临床研究 (2篇)

Progress in retinal and eye research IF 16.2 2026-11-11 PMID: 43014503
Acute zonal occult outer retinopathy (AZOOR) and the broader "AZOOR complex" have historically encompassed several disorders sharing acute visual symptoms and outer retinal dysfunction. Advances in multimodal imaging have revealed structural abnormalities in many of these presentations, challenging the original concept of an "occult" retinopathy without a visible anatomical correlate. We therefore revisit these concepts in light of modern imaging and current understanding of outer retinal disease. A systematic reassessment of 207 eligible published cases showed that 188 (90.8%) could be assigned a more specific contemporary diagnostic category based on the information available in the original reports. The most frequent assignments were multiple evanescent white dot syndrome (MEWDS), acute outer retinopathy (AOR), and multizonal outer retinopathy and retinal pigment epitheliopathy (MORR). These findings do not establish that the original diagnoses were incorrect or that each contemporary diagnosis represents a biologically independent disease, but show that many presentations historically grouped under AZOOR can now be characterized more specifically. We review the clinical features, multimodal imaging findings, natural history, and management of the principal retinal entities identified through this reassessment. We also examine inherited, toxic, autoimmune, paraneoplastic, post-infectious, and other disorders that may produce similar retinal appearances or provide the underlying clinical or etiological setting for these presentations. Rather than proposing a new disease classification, we suggest a pattern-based diagnostic approach in which OCT and fundus autofluorescence form the core imaging assessment, with additional imaging modalities selected according to the clinical presentation and differential diagnosis. Imaging is used to characterize the predominant pattern and anatomical level of retinal injury and, when possible, to identify a more specific retinal diagnosis; clinical history, systemic evaluation, laboratory testing, electrophysiology, and genetic testing can then guide further investigation of its etiological setting or identify a specific alternative diagnosis. This approach allows patients presenting within the broad clinical spectrum historically associated with AZOOR to be characterized at the most specific level supported by contemporary imaging, while recognizing that the biological relationships among these retinal entities remain incompletely understood.
中文摘要:急性区段隐匿性外视网膜病变(AZOOR)及其更广泛的「AZOOR复合征」在历史上涵盖了多种具有急性视觉症状和外视网膜功能障碍的疾病。多模态成像技术的进步揭示了这些表现中的许多存在结构异常,挑战了原本认为无可见解剖对应物的「隐匿性」视网膜病变概念。因此,我们结合现代成像技术和当前对外视网膜疾病的理解重新审视这些概念。对207例符合条件的已发表病例进行系统再评估显示,188例(90.8%)可根据原始报告中的信息被归入更具体的当代诊断类别。最常见的归类为多发性瞬态白点综合征(MEWDS)、急性外视网膜病变(AOR)以及多灶性外视网膜病变和视网膜色素上皮病变(MORR)。这些发现并未证明原始诊断错误或每个当代诊断代表一种生物学上独立的疾病,而是表明历史上归入AZOOR范畴的许多表现现在可以被更具体地描述。我们回顾了通过此再评估确定的主要视网膜实体的临床特征、多模态成像结果、自然病程和管理策略。我们还考察了可能产生相似视网膜外观或为这些表现提供潜在临床或病因背景的遗传性、毒性、自身免疫性、副肿瘤性、感染后及其他疾病。我们并非提出新的疾病分类,而是建议一种基于模式的诊断方法,其中光学相干断层扫描(OCT)和眼底自发荧光构成核心成像评估,其他成像模态根据临床表现和鉴别诊断进行选择。成像用于表征视网膜损伤的主要模式和解剖层次,并在可能时识别更具体的视网膜诊断;临床病史、全身评估、实验室检查、电生理学和基因检测则可指导对其病因背景的进一步调查或识别特定的替代诊断。这种方法允许将历史上与AZOOR相关的广泛临床谱系中的患者以当代成像支持的最具体水平进行表征,同时承认这些视网膜实体之间的生物学关系仍未完全明确。
Medical image analysis IF 14.0 2026-11-09 PMID: 43014078
Effective stroke prevention relies on identifying at-risk individuals before neurological symptoms arise, yet this is severely limited by the absence of scalable screening tools. To address this, we developed RetCEVD, a diagnostic framework that uses rapid, non-invasive retinal Optical Coherence Tomography Angiography (OCTA) for auxiliary assessment of ICVD. At its core, RetCEVD introduces a memory-guided mutual information optimization strategy that maximizes shared disease-relevant information within the same diagnostic category while suppressing redundant or confounding dependencies across different categories. This dual information constraint encourages compact, discriminative representations of subtle retinal microvascular abnormalities. Trained and validated on a multi-center cohort of 24,438 retinal images, RetCEVD addresses three primary tasks: auxiliary detection of existing ICVD, etiological subtyping (large artery atherosclerosis vs. small vessel disease), and current symptomatic-status classification (asymptomatic vs. symptomatic LAA). The model achieved AUCs of 0.95 internal and 0.94 external for ICVD detection, 0.83 internal and 0.81 external for etiological subtyping, and 0.75 internal and 0.76 external for current symptomatic-status classification. The two-year ALAA follow-up provided supplementary longitudinal validation for subsequent stroke events (AUC, 0.71). By comprehensively evaluating the model's clinical interpretability, including its learned pathological features and alignment with established clinical patterns, we provide novel insights into retina-based stroke evaluation. Ultimately, RetCEVD may serve as an efficient, low-cost screening and triage tool to identify individuals who may benefit from confirmatory neurovascular imaging and specialist assessment.
中文摘要:有效的卒中预防依赖于在神经系统症状出现前识别高危人群,但这一目标因缺乏可扩展的筛查工具而受到严重限制。为解决这一问题,我们开发了RetCEVD诊断框架,利用快速、非侵入性的视网膜光学相干断层扫描血管成像(OCTA)对缺血性心血管病(ICVD)进行辅助评估。RetCEVD的核心是一种记忆引导的互信息优化策略,该策略在最大化同一诊断类别内共享的疾病相关信息的同时,抑制不同类别间的冗余或混淆依赖关系。这种双重信息约束鼓励形成紧凑且具有判别力的视网膜微血管细微异常表征。基于包含24,438张视网膜图像的多中心队列进行训练和验证,RetCEVD完成了三项主要任务:现有ICVD的辅助检测、病因分型(大动脉粥样硬化与小血管病)以及当前症状状态分类(无症状与有症状LAA)。该模型在ICVD检测方面取得了0.95的内部AUC和0.94的外部AUC,在病因分型方面取得了0.83的内部AUC和0.81的外部AUC,在当前症状状态分类方面取得了0.75的内部AUC和0.76的外部AUC。为期两年的LAA随访提供了对后续卒中事件的补充纵向验证(AUC为0.71)。通过全面评估模型的临床可解释性,包括其学习的病理特征与既定临床模式的一致性,我们为基于视网膜的卒中评估提供了新见解。最终,RetCEVD可能作为一种高效、低成本的筛查和分诊工具,用于识别可能从确认性神经血管影像检查和专科评估中受益的人群。

基础研究 (2篇)

JAMA ophthalmology IF 10.5 2026-11-12 PMID: 43018943
Medical artificial intelligence (AI) can be trusted only if validated in patients not used for development. In glaucoma AI, a few public datasets from 1 laboratory serve as those independent cohorts. To determine whether 3 public glaucoma datasets used for external validation and fairness evaluation are independent patient cohorts. This cross-sectional dataset audit and record-linkage analysis was conducted in July 2026 using the publicly distributed files of Harvard-GDP (1000 eyes), Harvard-GF (3300 eyes), and Harvard-FairVision (10 000 eyes), all from a single laboratory. Records were linked by exact visual field and demographic fingerprints, with chance modeled from the within-dataset collision rate and permutation. All 1302 candidate pairs underwent direct image comparison. Number of shared eyes and image-level identity of optical coherence tomography (OCT) scans distributed under different identifiers. Harvard-GF and Harvard-FairVision matched on 1302 eyes against 1.9 expected by chance. Direct comparison confirmed 1031 (79.2%; 95% CI, 76.9%-81.4%) as the same 3-dimensional OCT volume (cosine similarity ≥0.9999; 88% of voxels exactly equal), at least 31% of the 3300 Harvard-GF eyes, a lower bound. Duplication was similar across the 3 Harvard-FairVision splits (78.5%-80.8%). Harvard-GDP and Harvard-GF shared 22 eyes matching exactly on all 52 total-deviation points and mean deviation, all in the Harvard-GF validation split; Harvard-GDP distributes no OCT volume, so its overlap rests on weaker evidence. No file carried a patient identifier, and no release disclosed the overlap; among the 64 of 105 citing studies whose full text we obtained, none reported it. Across 5 training runs, a classifier trained on 1 dataset and evaluated on the other gave an area under the curve of 0.90 (95% CI, 0.83-0.94) where 78.5% (95% CI, 74.0%-82.5%) of test eyes were training data, against 0.78 (95% CI, 0.72-0.83) out of sample (inflation, 0.12; 95% CI, 0.06-0.18). Inflation occurred in every racial subgroup and was larger where out-of-sample performance was lower (Spearman ρ = -0.57; permutation P = .02). These findings suggest that the examined datasets were not independent cohorts. Treating them as independent benchmarks inflates reported performance and might mask fairness gaps through data leakage and nonindependent validation.
中文摘要:医学人工智能仅在未用于开发的独立患者群体中验证方可信赖。在青光眼人工智能领域,少数来自单一实验室的公共数据集充当了这些独立队列的角色。本研究旨在确定用于外部验证和公平性评估的三个公共青光眼数据集是否为独立的患者队列。该横断面数据集审计与记录链接分析于2026年7月进行,使用了哈佛GDP(1000只眼)、哈佛GF(3300只眼)和哈佛FairVision(10000只眼)的公开分发文件,均源自单一实验室。通过精确视野和人口统计学指纹进行记录链接,并利用数据集内碰撞率和排列模拟机会概率。所有1302对候选对象均接受了直接图像比较。结果显示,在哈佛GF与哈佛FairVision之间,有1302只眼匹配,而预期随机匹配数仅为1.9只。直接比较证实其中1031只(79.2%;95%CI,76.9%-81.4%)为相同的三维光学相干断层扫描(OCT)体积(余弦相似度≥0.9999;88%的体素完全相等),这至少占哈佛GF 3300只眼的31%,为下限。在三个哈佛FairVision拆分中,重复率相似(78.5%-80.8%)。哈佛GDP与哈佛GF有22只眼在所有52个总偏差点和平均偏差上完全匹配,且均位于哈佛GF的验证拆分中;由于哈佛GDP未分发OCT体积,其重叠基于较弱的证据。没有任何文件包含患者标识符,也没有任何发布披露了这种重叠;在获取全文的105篇引用研究中的64篇中,无一报告此情况。在五次训练运行中,在一个数据集上训练并在另一个数据集上评估的分类器,当78.5%(95%CI,74.0%-82.5%)的测试眼为训练数据时,其曲线下面积为0.90(95%CI,0.83-0.94),而在样本外数据中为0.78(95%CI,0.72-0.83)(膨胀值为0.12;95%CI,0.06-0.18)。这种膨胀出现在每个种族亚组中,且在样本外性能较低时更为显著(Spearman ρ = -0.57;排列P = .02)。这些发现表明,所检查的数据集并非独立队列。将它们视为独立基准会夸大报告的性能,并可能通过数据泄露和非独立验证掩盖公平性差距。
Progress in retinal and eye research IF 16.2 2026-11-10 PMID: 43008845
Retinal imaging has traditionally described tissue and layer level pathology, whereas direct assessment of individual cells in the living human retina has been limited by ocular aberrations, optical contrast and image analysis constraints. Adaptive optics (AO) has substantially changed this landscape by improving lateral resolution and enabling in vivo visualization of photoreceptors, retinal pigment epithelium (RPE), ganglion-cell-layer structures, immune cells and microvasculature. This review discusses human retinal cellular imaging across three major AO-enhanced platforms: flood-illumination adaptive optics (FIAO), adaptive optics optical coherence tomography (AO-OCT) and adaptive optics scanning laser ophthalmoscopy (AOSLO). These modalities provide complementary capabilities, including rapid en face imaging, depth-resolved volumetric assessment and flexible confocal, non-confocal and fluorescence-based contrast. Recent progress is also being shaped by faster wavefront correction, wider-field acquisition, compressed multichannel imaging, sensorless AO and artificial intelligence (AI)-assisted reconstruction, segmentation and quantification. Beyond optical resolution, molecular labeling strategies are beginning to add biological specificity to retinal cellular imaging. Indocyanine green (ICG) supports AO-assisted assessment of RPE mosaic contrast, annexin V-based Detection of Apoptosing Retinal Cells (DARC) provides a molecular readout of retinal cell stress and apoptosis, and curcumin-based imaging has been explored as an experimental retinal amyloid marker in Alzheimer's disease. Together, these advances indicate a shift from high-resolution visualization alone towards cellular and molecular biomarkers of disease activity, progression and treatment response. Further translation will require standardized acquisition, robust automated analysis, multimodal validation and larger prospective studies across devices, disease populations and clinical sites.
中文摘要:视网膜成像传统上描述组织和层级的病理改变,而活体人眼视网膜单个细胞的直接评估受限于眼部像差、光学对比度和图像分析约束。自适应光学显著改变了这一局面,通过提高横向分辨率并实现视锥细胞、视网膜色素上皮、神经节细胞层结构、免疫细胞和微血管的体内可视化。本综述讨论了三种主要自适应光学增强平台上的视网膜细胞成像:泛光照明自适应光学、自适应光学光学相干断层扫描和自适应光学扫描激光检眼镜。这些模式提供互补能力,包括快速眼前面成像、深度分辨体积评估以及灵活的共聚焦、非共聚焦和基于荧光的对比度。最近的进展还受到更快的波前校正、更宽视野采集、压缩多通道成像、无传感器自适应光学和人工智能辅助重建、分割和量化的影响。除了光学分辨率外,分子标记策略开始为视网膜细胞成像增加生物学特异性。吲哚菁绿支持自适应光学辅助的视网膜色素上皮马赛克对比度评估,基于膜联蛋白V的凋亡视网膜细胞检测提供视网膜细胞应激和凋亡的分子读数,而基于姜黄素的成像已被探索作为阿尔茨海默病的实验性视网膜淀粉样蛋白标志物。这些进展共同表明从单纯高分辨率可视化向疾病活动、进展和治疗反应的细胞和分子生物标志物的转变。进一步转化需要标准化采集、稳健的自动分析、多模态验证以及跨设备、疾病人群和临床站点的大规模前瞻性研究。

3青光眼 (2篇)

临床研究 (2篇)

Ophthalmology IF 10.9 2026-11-12 PMID: 43020385
Open-angle glaucoma (OAG) is the leading cause of irreversible blindness and is highly heritable. Despite effective treatments, approximately 50% of cases remain undiagnosed. We aimed to develop and validate a clinically applicable, genetics-based risk prediction tool for OAG across diverse ancestries. Genetic association study and risk prediction analysis. More than 6 million individuals contributing genome-wide association study data from multiple ancestry groups. We constructed a comprehensive polygenic risk score (PRS) for OAG using a multitrait modeling framework that integrated glaucoma-related traits, including intraocular pressure, optic disc parameters, and ocular hypertension. PRS performance was evaluated for OAG risk across major ancestry groups and assessed for associations with age at diagnosis, treatment escalation, surgical outcomes, and familial risk. Odds ratios (ORs) for OAG by PRS percentile, age at disease onset, absolute risk estimates, treatment escalation, glaucoma surgery, and OAG risk in first-degree relatives. In Australian individuals of European ancestry, those in the top 10% of the PRS distribution had a 10-fold increased risk of OAG compared with the remainder (OR 10.04, 95% CI 7.43-13.59), substantially outperforming our previously published PRS (OR 4.2). Similar performance was observed in U.S. European ancestry data (OR 9.09, 95% CI 5.09-16.26). In UK Biobank Europeans the OR for the top 10% versus the remainder was 4.91 (95% CI 4.25-5.69). Although performance was attenuated in other ancestry groups (OR top 10% vs. remainder: 4.01 in South Asian, 3.39 in East Asian, and ranging from 2.26 to 3.81 in African ancestry populations depending on cohort), high PRS percentiles corresponded to elevated absolute risk across all groups. On average, individuals in the highest-risk decile developed glaucoma approximately 25 years earlier than those in the lowest decile and had ∼100-fold (OR 96.83, 95% CI 52.61-178.21) higher risk. The PRS also predicted treatment escalation in early disease, prevalent and incident glaucoma surgery, and OAG risk among first-degree relatives. This enhanced PRS, available for clinical use through an accredited genotyping workflow, enables effective risk stratification for OAG across diverse ancestries, facilitating earlier identification and treatment of high-risk individuals while supporting resource rationalisation for those at low risk. Performance is strongest in European ancestry populations, and further work is needed to improve equity of performance across ancestries.
中文摘要:开角型青光眼是导致不可逆失明的主要原因且遗传度极高。尽管治疗有效,约50%的病例仍未被诊断。本研究旨在开发并验证一种适用于临床、基于遗传学的开角型青光眼风险预测工具,涵盖不同祖先群体。研究类型为遗传关联研究与风险预测分析,数据来自多个祖先群体的超过600万参与者的全基因组关联研究。我们利用多性状建模框架构建了全面的开角型青光眼多基因风险评分,整合了眼压、视盘参数和眼高血压等青光眼相关性状。在多基因风险评分在主要祖先群体中的开角型青光眼风险评估中表现优异,并评估了其诊断年龄、治疗升级、手术结果及家族风险的关联。结果显示,在欧洲血统的澳大利亚个体中,多基因风险评分分布前10%的人群患开角型青光眼的风险是其余人群的10倍(OR 10.04,95% CI 7.43-13.59),显著优于我们此前发布的评分(OR 4.2)。美国欧洲血统数据也观察到类似表现(OR 9.09,95% CI 5.09-16.26)。在英国生物样本库欧洲人群中,前10%与其余人群的比值比为4.91(95% CI 4.25-5.69)。尽管在其他祖先群体中表现有所减弱(南亚为4.01,东亚为3.39,非洲血统群体因队列不同在2.26至3.81之间),但高多基因风险评分百分位在所有群体中均对应较高的绝对风险。平均而言,最高风险十分位人群比最低风险十分位人群早约25年发病,且风险高出约100倍(OR 96.83,95% CI 52.61-178.21)。该评分还能预测早期疾病的治疗升级、患病及新发青光眼手术以及一级亲属的开角型青光眼风险。这一增强版多基因风险评分通过认证的基因分型工作流程可用于临床,能够实现不同祖先群体中开角型青光眼的有效风险分层,促进高风险个体的早期识别与治疗,同时支持对低风险人群的资源合理化配置。该评分在欧洲血统人群中表现最强,仍需进一步工作以改善不同祖先群体间的性能公平性。
Science advances IF 13.9 2026-11-06 PMID: 42981387
Glaucoma, a leading cause of irreversible blindness, involves progressive retinal ganglion cell loss. Early detection of this neurodegeneration is critical, yet current clinical methods often identify disease only after substantial damage has occurred. This preliminary study used high-resolution adaptive optics-optical coherence tomography (AO-OCT) to track individual retinal ganglion cells in patients and healthy volunteers over 3 to 5 years. Results revealed higher neurodegeneration in glaucoma compared to normal aging. We observed 4.16 to 6.86% annual decrease in cell density in glaucoma, far exceeding the 0.27% cell loss rate observed in the healthy volunteer group. Critically, AO-OCT detected progression significantly earlier, identifying cell loss 0.74 years before conventional OCT and 1.49 years before visual field testing. These findings demonstrate AO-OCT as a more sensitive tool capable of monitoring glaucomatous damage with superior precision, potentially representing a major advancement over current clinical imaging.
中文摘要:青光眼是导致不可逆失明的主要原因,涉及进行性视网膜神经节细胞丢失。早期发现这种神经退行性变至关重要,然而当前的临床方法往往在造成严重损害后才识别出疾病。这项初步研究利用高分辨率自适应光学光学相干断层扫描(AO-OCT)追踪了患者和健康志愿者3至5年间的单个视网膜神经节细胞。结果显示,与正常衰老相比,青光眼患者的神经退行性变更为显著。我们观察到青光眼患者细胞密度每年下降4.16%至6.86%,远高于健康志愿者组观察到的0.27%细胞丢失率。关键的是,AO-OCT检测出进展的时间显著更早,比传统OCT早0.74年发现细胞丢失,比视野测试早1.49年。这些发现证明AO-OCT是一种更敏感的工具,能够以前所未有的精度监测青光眼损伤,可能代表当前临床成像技术的重大进步。

4眼肿瘤 (1篇)

临床研究 (1篇)

American journal of ophthalmology IF 4.7 2026-11-10 PMID: 43009085
To evaluate the outcomes and safety of high-dose intravitreal topotecan (100 μg, IVitT100) in the treatment of refractory or recurrent intraocular retinoblastoma (RB). Prospective, non-comparative case series. 15 consecutive patients (15 eyes) with refractory or recurrent RB at the Eye & ENT Hospital of Fudan University between October 2023 and September 2025. Participants received IVitT100 every four weeks. Clinical characteristics were recorded before and after treatment. Tumor control and injection-related complications were assessed at each visit. Aqueous humor cytokines and electroretinography (ERG) were analyzed to evaluate inflammatory changes and retinal function. Treatment outcomes, including intraretinal tumor and intraocular seeds control, globe salvage rate, and ocular and systemic adverse events, were assessed during follow-up. At a median follow-up of 9 months (IQR, 6.0-14.0) after the final intravitreal injection, high-dose intravitreal topotecan achieved a globe salvage rate of 93%, with complete control of intraocular seeds in all eyes (100%) and complete regression of intraretinal tumors in 3 of 4 eyes (75%). Baseline ERG amplitude was 5.7 μV (IQR, 1.5-14.7), post-treatment amplitude was 5.7 μV (IQR, 1.9-15.8), showing no significant retinal function impairment (P = 0.429). None of the 27 cytokines showed significant changes between pre- and post-treatment measurements in RB eyes (all P > 0.05). No severe ocular or systemic adverse events were reported. IVitT100 is highly effective for refractory or recurrent RB with vitreous seeds, subretinal seeds, and intraretinal tumors. It exhibits minimal retinal toxicity and does not induce intraocular inflammatory responses, supporting its role as a safe and effective first-line intravitreal therapy for advanced RB.
中文摘要:本研究旨在评估高剂量玻璃体内注射拓扑替康(100 μg,IVitT100)治疗难治性或复发性眼内视网膜母细胞瘤(RB)的疗效与安全性。这是一项前瞻性、非对照病例系列研究,纳入2023年10月至2025年9月期间复旦大学附属眼耳鼻喉科医院的15例连续患者(15只眼)。参与者每四周接受一次IVitT100治疗。研究记录了治疗前后的临床特征,并在每次随访时评估肿瘤控制情况及注射相关并发症。通过分析房水细胞因子和视网膜电图(ERG),评估炎症反应变化及视网膜功能。在最后一次玻璃体内注射后中位随访9个月(IQR,6.0-14.0)期间,高剂量玻璃体内注射拓扑替康的眼球保留率达93%,所有眼球(100%)的玻璃体种子细胞完全控制,4只眼中有3只眼(75%)的视网膜内肿瘤完全消退。基线ERG振幅为5.7 μV(IQR,1.5-14.7),治疗后振幅为5.7 μV(IQR,1.9-15.8),显示无显著视网膜功能损害(P = 0.429)。RB眼在治疗前后27种细胞因子均未出现显著变化(所有P > 0.05)。未报告严重的眼部或全身不良事件。IVitT100对伴有玻璃体种子、视网膜下种子及视网膜内肿瘤的难治性或复发性RB具有高度疗效,表现出极低的视网膜毒性且不诱导眼内炎症反应,支持其作为晚期RB安全有效的首选玻璃体内治疗手段。

5神经眼科 (1篇)

基础研究 (1篇)

EBioMedicine IF 11.2 2026-11-10 PMID: 43008336
Progressive supranuclear palsy (PSP) is a neurodegenerative disorder characterised by a combination of motor, ocular, cognitive, and behavioural impairment, but has a heterogeneous clinical presentation. It is a devastating and rapidly progressive atypical parkinsonian disorder with no current disease-modifying treatments. PSP is classified as a primary tauopathy characterised by abnormal accumulation of tau within neurons and glial cells, primarily in the brain stem, culminating in cell death. While genome wide association studies have been instrumental in identifying multiple risk loci, recent multi-omics studies combining brain gene expression and other omics measures have revealed a plethora of perturbed transcripts and molecular pathways that may play pivotal roles in PSP. This includes genes such as MAPT, STOM, KANK2, DDR2, LRRK2, and STX6, but their precise roles remain elusive. These genes are differentially expressed and may regulate processes including neuroinflammation, glymphatic clearance of tau, cytoskeletal maintenance, and synaptic function. In this review, we summarise current treatment strategies attempted in PSP, compile disease-modifying therapies in development, and explore potential disease mechanisms suggested by multi-omics studies. By targeting several underlying disease processes at once, combination strategies have the potential to improve both motor and cognitive function, slow neurodegeneration, and ultimately alter disease progression.
中文摘要:进行性核上性麻痹(PSP)是一种以运动、眼部、认知和行为障碍为特征的神经退行性疾病,临床表现具有异质性。这是一种毁灭性且快速进展的非典型帕金森综合征,目前尚无疾病修饰疗法。PSP被归类为主要由tau蛋白异常积聚引起的原发性tau蛋白病,主要发生在脑干,最终导致细胞死亡。虽然全基因组关联研究在识别多个风险位点方面发挥了重要作用,但最近结合脑部基因表达和其他组学指标的多组学研究揭示了众多受扰转录本和分子通路,这些可能在PSP中发挥关键作用。其中包括MAPT、STOM、KANK2、DDR2、LRRK2和STX6等基因,但其确切作用仍不明确。这些基因差异表达并可能调节神经炎症、tau蛋白的类淋巴清除、细胞骨架维持和突触功能等过程。在本综述中,我们总结了目前尝试的治疗策略,汇编了正在开发的疾病修饰疗法,并探讨了多组学研究提出的潜在疾病机制。通过同时针对多种潜在的疾病过程,组合策略有望改善运动和认知功能,减缓神经退行性变,并最终改变疾病进程。

6角膜与眼表疾病 (1篇)

基础研究 (1篇)

Progress in retinal and eye research IF 16.2 2026-11-09 PMID: 43003180
Slit-lamp colour photography remains a cornerstone of anterior segment documentation and longitudinal monitoring. However, conventional slit-lamp photography has seen relatively limited technological advancement compared with the rapid development of quantitative anterior-segment modalities such as AS-OCT, Scheimpflug tomography and confocal microscopy. Despite the wide diffusion of digital cameras, smartphones, and high-end slit-lamp systems, ocular surface photography (OSP) remains hampered by the absence of standardised acquisition and processing protocols. The convex, aspheric geometry of the ocular surface, off-axis illumination, peripheral defocus, automatic colour correction, and pupil-driven exposure variation in retroillumination limit reproducibility, inter-device comparability, and clinical utility. In this review we critically appraise current OSP applications across major ocular surface conditions (dry eye disease, conjunctival hyperaemia, infectious keratitis, ocular surface burns, and ocular surface neoplasia), mapping the hardware, software, and reproducibility limitations that constrain clinical practice. We then examine how these limitations propagate into artificial intelligence (AI) development, where the absence of large, well-annotated, demographically diverse OSP datasets, together with the lack of reproducible quantitative phenotypes and reliable labels, is now the principal bottleneck to clinically deployable models; we discuss patient-level data partitioning and external validation, and outline how deep learning, federated learning and foundation models can both utilise and improve OSP. Finally, we propose a minimum acquisition and dataset specification, tiered by clinical setting from research to community screening, and a staged roadmap that combines hardware standardisation (focus-stacked, illumination-controlled, colour-calibrated acquisition), interoperable metadata-rich storage (DICOM where infrastructure permits), quantitative phenotype definition, and diverse international dataset collaborations, so that the ocular surface can benefit from quantitative imaging and AI as other areas of ophthalmology have begun to.
中文摘要:裂隙灯彩色摄影仍是眼前段记录和纵向监测的基石。然而,与AS-OCT、Scheimpflug断层扫描和共聚焦显微镜等定量眼前段模态的快速进展相比,传统裂隙灯摄影的技术进步相对有限。尽管数码相机、智能手机和高端裂隙灯系统已广泛普及,但眼表摄影(OSP)仍因缺乏标准化的采集和处理协议而受到阻碍。眼表的凸面非球面几何形状、离轴照明、周边散焦、自动色彩校正以及瞳孔驱动的逆光曝光变化限制了可重复性、设备间可比性和临床实用性。本综述批判性地评估了当前OSP在主要眼表疾病(干眼病、结膜充血、感染性角膜炎、眼表烧伤和眼表肿瘤)中的应用,映射了限制临床实践的硬件、软件和可重复性局限性。随后,我们探讨了这些局限性如何影响人工智能(AI)的发展,其中缺乏大规模、标注良好、人口统计学多样化的OSP数据集,以及缺乏可重复的定量表型和可靠标签,已成为部署临床模型的主要瓶颈;我们讨论了患者层面的数据划分和外部验证,并概述了深度学习、联邦学习和基础模型如何利用和改进OSP。最后,我们提出了最低限度的采集和数据集规范,按从研究到社区筛查的临床环境分层,并制定了分阶段路线图,结合硬件标准化(焦点堆叠、照明控制、色彩校准采集)、互操作性元数据丰富存储(在基础设施允许时使用DICOM)、定量表型定义和多样化的国际数据集合作,从而使眼表能够像眼科其他领域一样受益于定量成像和AI。