您的位置: 首页 > 2024年6月 第39卷 第6期 > 文字全文
2023年7月 第38卷 第7期11
目录

眼中洞见:人工智能解码全身健康

Insights from the eye: artificial intelligence decodes systemic health

来源期刊: 眼科学报 | 2024年6月 第39卷 第6期 317-324 发布时间:2024-06-28 收稿时间:2024/8/28 16:37:21 阅读量:930
作者:
关键词:
人工智能眼科AI心血管健康神经系统健康衰老
artificial intelligence ophthalmic AI cardiovascular health nervous system health aging
DOI:
10.12419/24062404
收稿时间:
2024-03-28 
修订日期:
2024-04-15 
接收日期:
2024-05-05 
人工智能(artificial intelligence, AI)在医学领域的广泛应用为探索眼部与全身健康的关系提供了新的机遇。文章回顾了眼科AI在心血管健康、神经系统健康、肾脏健康和衰老过程中的应用。在心血管健康方面,AI能够通过分析眼底图像预测心血管疾病风险因素和未来心血管事件,并提供了简便、有效的风险分层方法。在神经系统健康方面,眼科AI在阿尔茨海默病早期诊断和帕金森病识别方面显示出潜力,尽管对未来事件预测仍具挑战性。针对多发性硬化,眼科AI在诊断和预测残疾进程上展现了良好效果。在肾脏健康中,眼科AI技术通过分析视网膜图像可预测肾功能相关指标、直接检测肾病事件,展示了其在改善肾病筛查方式和减轻医疗负担方面的潜力。在衰老过程中,AI能够利用眼部图像预测生物年龄、视网膜年龄差和晶状体年龄等参数提供了生物衰老指标,为理解衰老与眼部健康的关联提供了新视角。
The widespread application of artificial intelligence (AI) in the medical field has provided new opportunities to explore the relationship between eye and whole body health. This article reviews the application of ophthalmic AI in cardiovascular health, neurological health and aging. In terms of cardiovascular health, AI can predict cardiovascular disease risk factors and future cardiovascular events by analyzing fundus images, and provides a simple and effective risk stratification method. In terms of neurological health, ophthalmic AI shows potential in early diagnosis of Alzheimer's disease and identification of Parkinson's disease, although the prediction of future events remains challenging. For multiple sclerosis, ophthalmic AI has shown good results in diagnosing and predicting the progression of disability. In kidney health, ophthalmic AI technology can predict kidney function-related parameters and detect kidney disease events by analyzing retinal images, demonstrating its potential in improving kidney disease screening methods and reducing medical burdens. In the aging process, AI can use eye images to predict biological age. Parameters such as retinal age gap and LensAge provide biological aging indicators, providing a new perspective for understanding the relationship between aging and eye health.

文章亮点

1. 关键发现

人工智能 (AI) 通过分析眼部图像能够有效检测全身疾病并预测未来发病风险。这一技术为全身健康的实时监控和早期干预提供了一种简便、无创且经济的新方法,并为理解眼与全身健康的关联提供了新视角。

2. 已知与发现

AI技术在医学领域的应用日益广泛。眼科AI不仅在眼部疾病检测上表现出色,还在评估全身健康方面取得了重要进展。目前的眼科 AI 可多方位提供关于心血管、神经系统、肾脏健康及衰老相关信息,显示了其在多个医学领域的应用潜力。

3. 意义与改变

通过对全身疾病的早期筛查和未来风险预测,眼科 AI 有望减轻医疗系统的负担,提高早期检测和干预能力。同时,眼科 AI 还提供了新的生物衰老指标,有助于深入研究衰老过程及其对健康的影响。这一技术推动了 AI 在眼科和其他医学领域的跨学科应用,拓展了其在医学中的应用范围。

       医学中的人工智能(artificial intelligence, AI)能够将机器学习模型应用于分析医疗数据,在临床环境和研究中为医疗专业人员提供宝贵支持,改善医疗结局和患者体验。医学AI涵盖了机器学习、深度学习和计算机视觉等多个领域。AI系统通过复杂的算法,如卷积神经网络、循环神经网络和强化学习,从大量的医疗数据中学习,具有自我纠正和根据反馈提升准确性的能力,协助医生进行诊疗和疾病推理,特别是在处理非结构化数据和复杂模式识别方面表现出色,还有助于减少人类临床实践中不可避免的诊断和治疗错误。
       在当前的医疗环境中,AI最常见的应用包括临床决策支持和医学影像分析。临床决策支持工具可帮助医疗保健提供者进行治疗决策,快速获取相关信息和研究成果。在医学影像领域,AI具有在图像中提取特征的优势,在图像分析中发挥了重要作用。目前已经有多种基于深度学习的系统可以分析视网膜图像,以辅助诊断和监测眼部疾病。较为成熟的算法包括通过微血管瘤和视网膜出血检测糖尿病性视网膜病变、通过脉络膜新生血管等病变检测晚期年龄相关性黄斑变性,以及通过视神经损伤判断青光眼[1]。AI的另一个重要优势是其快速处理图像的能力。AI算法可以在短时间内分析大量视网膜图像,实现高通量筛查和人群水平的分析。通过自动化的初步风险评估过程,AI可以帮助减轻医疗系统的负担,扩大筛查范围,并确保有限医疗资源的有效分配。此外,AI技术还可采用数据增强和迁移学习等方法来提高模型的鲁棒性和准确性,针对性解决医学影像分析中的数据噪声和样本不平衡问题[2-3]
       眼被认为是全身健康状况的窗户。通过非侵入性成像技术,我们能够从视网膜中直接观测到神经血管系统的末端,从而反映全身健康的状况。由于解剖和生理特性上与大脑和肾等末梢器官存在高度相似性,因此可将视网膜血管定位为全身微血管系统的镜像。在视网膜中检测到的早期微血管变化通常预示着包括卒中和缺血性心脏病在内的全身循环系统疾病[4]。作为中枢神经系统的延伸,视网膜与影响大脑和脊髓的神经退行性疾病也有着紧密的联系。已有大量研究表明,视网膜神经纤维层厚度和视力的变化与早期认知损害相关[5]。除此之外,外眼作为早期即出现黄疸、胆固醇沉积和贫血相关临床表现的主要部位,进一步反映了其他全身系统的健康状况。
       以技术为驱动的眼科领域,特别是其以图像为中心的特性,极大地促进了可用于开发AI算法的图像数据的积累。因此,眼科图像与AI分析的融合在预测眼科健康以及全身生物标志物和疾病方面具有巨大潜力,代表着一个蓬勃发展的研究领域。得益于批处理和并行计算的能力,AI在处理大规模数据和复杂模式识别方面表现出色。通过分析大量的视网膜图像和临床资料,AI能够识别可能导致疾病进展的模式和风险因素[6-7]。这种预测能力有助于临床医生制定早期干预方案,并为患者提供个性化治疗。本综述将探讨AI技术在探索眼与全身健康关联方面的现状与进展,包括在心血管健康、神经系统健康、肾脏健康及衰老等领域的应用和成就。

1 眼科AI在心血管健康中的应用

       眼部血管与心血管循环系统具有许多相似之处,并经常受到相同的内在和环境影响。因此,视网膜为我们提供了用于检测心血管疾病(cardiovascular diseases, CVD)微血管变化的独特窗口。视网膜微血管代表着全身循环系统的末梢。当视网膜血管存在病理变化,如视网膜动脉变窄或血管异常弯曲时,通常意味着整个循环系统中更广泛的损害。这些变化不是局限于眼部的孤立事件,而是长期存在CVD等问题造成的全身性损害的反映。多项大型人群研究揭示了视网膜血管与CVD风险之间的密切关联。ARIC研究、Rotterdam研究和多项流行病学调查表明,较小的动脉-静脉比(arterio-venous ratio, AVR)与更高的CVD风险密切相关[8-13]。因为视网膜血管位于复杂循环系统网络的末端,它的变化可作为体内血管整体健康的指标。因此,对视网膜生物标志物的分析为我们理解循环系统整体状况和其累积损伤提供了宝贵的窗口。

1.1 预测CVD风险因素

       计算机科学和信息学的进步促使AI迅速融入现代医疗保健,为临床环境和研究中的医疗专业人员提供了宝贵的支持。受益于AI的发展,我们可以从眼底图像中提取和分析海量的解剖学特征。特别是深度学习算法等前沿技术在预测CVD风险因素方面提供了可靠的结果。利用广泛数据集训练的深度学习模型能够利用解剖特征(如视盘、视网膜血管),从而预测年龄、性别甚至种族等CVD相关的人口特征。现有的AI模型已经能够达到较高的准确性,从眼底图像推断的相关性或曲线下面积(area under the curve, AUC)范围为0.85~0.97,年龄预测的平均绝对误差约为3岁[7,14-15]。同样,AI也能通过光学相干断层扫描(optical coherence tomography, OCT)图像识别CVD相关的人口特征,例如对年龄预测的平均绝对误差为5~6岁,性别预测的AUC范围为0.72~0.96[16-17]。一些具有更高准确性和针对性的算法能够通过OCT图像中黄斑中心凹的特征来预测性别,通过整个视网膜层的特征进行年龄预测[18]
       此外,深度学习算法还展示了预测CVD相关临床信息的能力。新加坡的一项研究评估了这些算法在预测各种系统生物标志物方面的潜力,包括身体成分测量、血压和血液学参数[19]。然而,现有算法在预测连续变量的临床因素方面表现普遍较差,相关性难以达到0.6。此外,这些模型在不同族裔数据集上的泛化能力也较差,未来仍需要未来在不同背景的人群中进行交叉验证。通过深度学习技术,还可以从眼底图像推断与生活方式相关的CVD风险因素,如吸烟和饮酒状况[20]。然而在这些研究中,生活方式状态往往基于患者自我报告,可能受到回忆或信息偏差的影响。

1.2 CVD风险分层和未来心血管事件预测

       准确的风险分层对于识别和管理CVD风险个体至关重要。目前有多种临床常用的CVD风险计算公式,包括Pooled cohort equation、Framingham score和QRISK3等,可用于估计未来CVD的风险[21-22]。目前筛查CVD风险的计算公式通常依赖于多种临床信息,包括年龄、性别、吸烟状况、血压、体质量指数、血糖水平和血胆固醇水平。然而,在所有患者中获得胆固醇水平等侵入性血液检测的数据仍然存在挑战,特别是在医疗资源分配不均的地区[23]
       深度学习算法可以利用眼底图像作为简单输入进行CVD风险分层。Joseph等[24]引入了一种基于深度学习的CVD视网膜生物标志物,称为Reti-CVD,旨在识别中度和高度CVD风险的个体,并显示出与现有临床CVD风险评分的高度可比性。此外,利用全身及视网膜血管健康的参数,包括颈动脉粥样硬化、冠状动脉钙化、动脉硬度和视网膜血管直径等因素,可以进一步增强或提供个体化CVD风险预测的附加价值。成熟的深度学习算法通过眼底图像分析提供了一种自动、快速和非侵入性的手段来提取这些参数,并建立了预测值与CVD结果之间的稳健关联[25]。除了预测CVD风险分层外,研究人员还致力于通过眼底图像直接预测未来的CVD疾病事件。Poplin等[20]开发的深度学习模型能利用眼底图像预测未来5年内的重大不良心血管事件,并实现了类似于系统性冠状动脉风险评估量表(SCORE Risk Charts)的准确性(AUC为0.7)。本研究团队提出了“视网膜年龄差”这一概念,即基于眼底图像预测的视网膜年龄与实际年龄之间的差异。“视网膜年龄差”已被证明是一个有效的心脏健康事件预测生物标志物,能够预测包括动脉硬度、新发CVD和CVD相关死亡在内的一系列CVD相关结局[26]

2 眼科AI在神经系统健康中的应用

       视网膜与间脑在胚胎发育上有关,其在血管、血液屏障和病理生理学上与大脑具有结构上的相似性,故将其定义为中枢神经系统的延伸部分。哺乳动物实验揭示了一种从视网膜起始的细胞网络的存在,这一网络经过各种交互作用,最终到达大脑视皮层[27]。眼睛中存在着涵盖视网膜病变、视网膜和脉络膜层厚度、血管参数和视网膜电图(electror etinogram, ERG)指数等多种生物标志物,如何整合这些丰富信息以有效早期检测中枢神经系统疾病具有挑战,而AI则为此提供了客观手段和有望解决方案。

2.1 阿尔茨海默病的早期诊断

       监督式机器学习能够利用全面的视网膜参数,在阿尔茨海默病(Alzheimer's disease, AD)的临床诊断方面开辟了全新的道路。Zhang等[28]运用机器学习技术对眼底彩照进行血管分割,在对轻度认知障碍的检测中表现出不错的性能。Tian等[29]利用模块化机器学习技术与英国生物库数据,针对视网膜血管进行分析,并成功地在鉴别AD患者与正常个体方面达到82.44%的准确率 。将深度学习与视网膜成像结合应用于识别系统性疾病,包括AD及其临床前期形式是当前的研究热点。Zhang等[30]利用回顾性的多中心数据开发了一种深度学习算法,成功地从视网膜图像中检测了AD相关性痴呆。该算法的AUC可达0.73~0.91,并能够确定具有风险因素的个体(如β淀粉样蛋白阳性)。与临床数据相比,利用深度学习对视网膜图像进行分析和评估能够更准确地进行疾病状态判断。一项比较了多模态数据[包括视网膜图像、OCT、光学相干断层扫描血管造影(optical coherence tomography angiography, OCTA)定量数据和临床资料]的深度学习模型研究发现,神经节细胞内丛状层图像在AD识别中能够提供最为丰富的信息[31]。这些研究突出了基于AI的眼部图像分析作为AD检测的一个有希望的途径,其准确性与传统方法(如简易精神状态检查表)相当,但具有更简单和标准化的临床工作流程[32-33]
       尽管在从视网膜图像中识别常见的痴呆或AD方面取得了进展,但在利用AI预测未来疾病发生方面的证据仍然有限。视网膜特征与AD风险之间的纵向联系需要进一步探索。Hui等[34]提出了深度强化学习的实用性——一种强化学习和深度学习的融合的新技术,有望作为未来医学图像分析中高维观测的有力工具。随着AI技术的进步,未来的研究可以深入探讨视网膜成像与大脑中与AD进展相关的类似变化。

2.2 眼与帕金森病

       视网膜纹理能够反映视网膜层结构排列的特点,利用AI技术可从视网膜图像中检测帕金森病(Parkinson's disease, PD)的疾病状态。Nunes等[35]设计了一种模型,利用OCT图像中的纹理指标,即根据视网膜组织的结构排列来区分AD、PD和健康对照组。Reiner等[36]则创新地将计算机视觉和深度学习算法相结合,发现了眼动障碍与PD严重程度之间的相关性。他们的发现显示,随着PD的严重程度增加,眼动障碍程度也会进一步加重,表现为注视延迟时间延长和反应准确度降低。然而,直接从视网膜图像预测PD仍然具有挑战性。Ahn等[37]提出了一种基于深度学习的方法,旨在利用眼底图像和患者人口统计数据来预测PD中的神经功能障碍的情况。虽然这种方法在外部验证中仅取得了0.67的AUC和70.45%的准确率,低于可供临床使用的阈值,但此次探索仍为我们通过AI眼部评估进行非侵入性PD监测提供了前景。
       从视网膜图像直接预测未来的PD发展仍缺乏实质性、突破性的证据。我们的团队分析了来自35 834名参与者的数据并发现“视网膜年龄差”每增加1年,与PD风险增加10%相关。这突显了“视网膜年龄差”对未来PD发展的预测潜力,标志着眼科大量生物标志物中未开发的巨大潜力[38]

2.3 人工智能和多发性硬化的关联

       机器学习方法在整合多个OCT测量参数以用于诊断多发性硬化(multiple sclerosis, MS)和预测MS相关残疾进展方面已表现出了良好的效果。Montolío等[39]的研究比较了各种机器学习算法在基于临床数据和视网膜神经纤维层厚度诊断MS及预测其长期残疾进程方面的表现 。几乎所有的机器学习算法都具有不错的表现,其中集成分类器在MS诊断中最为合适,能够达到0.877 5的AUC;相比其他各类算法,长短期记忆循环神经网络则在预测远期MS相关残疾进展方面表现出最高的预测性能(AUC为0.816 5)。
       AI技术可以用于识别MS诊断的关键参数。López-Dorado等[40]利用OCT图像训练了一个诊断MS的神经网络模型,在目前的临床诊断标准指导下能够达到90%以上的准确性。他们的研究提出OCT测得的神经节细胞层和全视网膜厚度是MS诊断中最具区别性的特征。此外,Cavaliere等[41]结合了视网膜和脉络膜厚度数据,训练的SVM分类器对MS的检测准确性能达到AUC 0.97。然而,目前通过眼科AI技术对MS进行诊断仍然依赖于测量OCT参数作为中间步骤,AI主要作为整合各项参数及识别关键参数的角色发挥作用。未来的研究可考虑进一步利用深度学习的技术,构建眼与神经系统健康整体关联的基础模型,并进一步针对下游具体疾病,如MS的诊断与预测进行优化,提供更简单、直接、全局的临床决策支持。

3 眼科AI在肾脏健康中的应用

       尽管肾和眼看似功能迥异,但它们在发育途径和分子结构上存在相似之处。从组织胚胎学角度来看,肾和眼的器官发生期均为妊娠的第4~6周,两者有相似的细胞、血管结构与生成过程,且血流受相同的调节系统控制[42-43]。视网膜病变与肾功能障碍之间存在显著关联。例如,CRIC研究、NHANES研究和新加坡眼病流行病学研究均发现,肾功能损伤可增加约2~3倍的视力损伤和视网膜病变风险[44-46]。此外,视网膜血管征象如视网膜静脉扩张,与慢性肾脏病风险增加1.2倍有关,且可作为终末期肾病患者全因死亡率的强独立预测指标[47]。基于OCT和OCTA的高分辨率成像发现,视网膜神经纤维层厚度下降、视网膜毛细血管密度降低、脉络膜变薄等定量改变均提示肾功能异常,可作为肾功能下降的潜在生物标志物。
       利用眼和肾脏之间的密切关联的A I技术正在起步阶段。目前的主流研究可分为通过视网膜图像预测肾功能相关指标,如肾小球滤过率(estimated glomerular filtration rate, eGFR),和直接预测肾衰竭相关事件两种类型。Kang等[48]尝试利用深度学习技术通过眼底彩照检测早期肾衰竭[定义为eGFR < 90 mL/(min·1.73 m2)]。尽管该模型能够达到0.81的总体AUC,但特异度较差,仅为0.6。尤其在伴随其他眼底病变,如视网膜瘢痕或视盘水肿的病例中,该模型倾向于给出假阳性的预测结果,临床实用性可能有限。Rim等[19]对使用眼底照片预测不同系统生物标志物进行了更广泛的研究 ,可以从眼底彩照中精确估算血肌酐水平,相关系数可高达0.72~0.82。
       另一些AI算法则专注于直接检测肾病事件。Betzler等[49]开发的深度算法可从糖尿病患者的视网膜照片中检测糖尿病肾病,结合视网膜图像与临床信息时模型能够达到最佳准确性(AUC为0.761~0.866)。Wong等[50]开发的深度学习模型针对慢性肾脏病进行直接检测,在仅基于眼底彩超进行预测时表现出接近传统临床肾衰竭风险评估的预测能力(AUC 0.911 vs. 0.916),提示眼底彩照可作为现有血液指标筛查的无创替代方案 。本团队的研究还发现,基线“视网膜年龄差” 每增加1年,10年内的独立肾衰竭风险则增加10%[51]。目前的研究主要集中于从眼底彩照中进行基于AI的肾功能参数估算和肾病预测,这些方法有望大幅减少需要采集的生物样本数量,减少患者的医疗负担和成本,有助于早诊断、早干预。
       理论上,也可以通过 OCT 或 OCTA 图像对肾脏相关结果进行基于AI的识别,然而,该领域的相关研究却并不多。Liu等[52]开发的深度学习算法可从糖尿病患者的视网膜 OCT 图像中筛查出患有糖尿病肾病的高危人群,表现出良好的诊断性能(准确率91.68%)。未来研究可扩大模型训练数据量,进一步探索 OCT及OCTA在肾功能及肾病事件预测中的作用,关键在于区分眼底的细微改变是来源于肾功能不全抑或来自其他眼部或全身疾病。

4 眼科AI在衰老过程中的应用

       衰老过程深刻影响每个人器官的结构和功能。眼在衰老过程中经历了许多解剖学上的变化,包括细胞丢失、减弱、组织退化和代谢废物的积累。了解眼在衰老过程中的复杂变化,对揭示更广泛的生理性衰老模式具有重要潜力。对这些眼部变化进行早期跟踪提供了在年龄相关疾病发作之前可视化衰老效应的机会,也有助于早期发现加速衰老过程,并促进及时地治疗干预。
       Chueh等[18]构建了一种深度学习模型,可利用OCT图像在健康人群中判断性别和年龄。通过分析黄斑OCT图像,他们的模型在预测年龄方面达到了较高的准确性,平均绝对误差约为5.78岁。外眼照片也包含了预测年龄的信息。先前的研究表明,训练于内外眦和眼前段图像的深度学习模型在年龄预测方面具有不错的表现(平均绝对误差为2.3岁和2.8岁)[53-54]。虽然深度学习展示了在预测年龄方面的潜力,但其预测值与实际年龄之间的固有偏差通常在3~5岁。有一种猜测认为该类深度学习模型从眼底图像中所得的预测值可能反映了个体的生物年龄,而这种预测年龄与实际年龄之间的差异或许表明了个体加速或减缓的衰老过程,强调了寻找生物衰老的有效指标的重要性。
       Nusinovici等[55]采用一种名为“RetiAGE”的深度学习算法,预测个体≥65岁的概率。较高的RetiAGE预测值与全因死亡和疾病相关死亡风险增加,提示RetiAGE能够反映个体内在的生物年龄。本研究团队所定义的“视网膜年龄差”作为衡量生物衰老的新指标,其预测值每增加1年与全因死亡风险增加2%,以及与非心血管和非癌症相关疾病致死风险增加3%相关[56]。除视网膜之外,晶状体也是反映衰老改变的重要窗口。Li等[57]利用晶状体累积的衰老相关变化,开发了基于眼前段照片预测实际年龄的深度学习模型,引入了新的参数“lensAge指数”。通过对比晶状体年龄和实际年龄,“lensAge指数”能够反映个体相对于同龄人的生物学衰老速率,有效地揭示了与衰老相关的眼部疾病、全身疾病和死亡风险。

5 结论

       AI在眼科领域展示了在预测全身健康和疾病风险方面的潜力。在心血管健康方面,AI技术有望成为快速、非侵入性的心血管风险预测工具,但仍需解决泛化性和连续变量预测的挑战。在神经系统健康方面,AI在AD、PD和MS的诊断中表现出了潜力,但对未来事件预测的证据仍有限。通过分析眼底彩照和OCT图像,AI在预测肾功能相关指标和检测肾病事件方面表现出色,尤其在慢性肾脏病的筛查中具有重要应用前景。此外,AI提供了新的生物衰老指标,如视网膜年龄差和晶状体年龄,为研究衰老与眼部健康关系提供了重要线索。然而,AI技术的应用仍需要更多研究来验证其在临床实践中的有效性和可靠性,以更好地推动医学领域的发展。

利益冲突

所有作者均声明不存在利益冲突。

开放获取声明

本文适用于知识共享许可协议 ( Creative Commons),允许第三方用户按照署名(BY)-非商业性使用(NC)-禁止演绎(ND)(CC BY-NC-ND)的方式共享,即允许第三方对本刊发表的文章进行复制、发行、展览、表演、放映、广播或通过信息网络向公众传播,但在这些过程中必须保留作者署名、仅限于非商业性目的、不得进行演绎创作。详情请访问:https://creativecommons.org/licenses/by-nc-nd/4.0/。




1、Ting DSW, Pasquale LR, Peng L, et al. Artificial intelligence and deep learning in ophthalmology[ J]. Br J Ophthalmol, 2019, 103(2): 167- 175. DOI: 10.1136/bjophthalmol-2018-313173.Ting DSW, Pasquale LR, Peng L, et al. Artificial intelligence and deep learning in ophthalmology[ J]. Br J Ophthalmol, 2019, 103(2): 167- 175. DOI: 10.1136/bjophthalmol-2018-313173.
2、Kebaili A, Lapuyade-Lahorgue J, Ruan S. Deep learning approaches for data augmentation in medical imaging: a review[ J]. J Imaging, 2023, 9(4): 81. DOI: 10.3390/jimaging9040081.Kebaili A, Lapuyade-Lahorgue J, Ruan S. Deep learning approaches for data augmentation in medical imaging: a review[ J]. J Imaging, 2023, 9(4): 81. DOI: 10.3390/jimaging9040081.
3、Kim HE, Cosa-Linan A, Santhanam N, et al. Transfer learning for medical image classification: a literature review[ J]. BMC Med Imaging, 2022, 22(1): 69. DOI: 10.1186/s12880-022-00793-7.Kim HE, Cosa-Linan A, Santhanam N, et al. Transfer learning for medical image classification: a literature review[ J]. BMC Med Imaging, 2022, 22(1): 69. DOI: 10.1186/s12880-022-00793-7.
4、Liew G, Gopinath B, White AJ, et al. Retinal vasculature fractal and stroke mortality[ J]. Stroke, 2021, 52(4): 1276-1282. DOI: 10.1161/ STROKEAHA.120.031886.Liew G, Gopinath B, White AJ, et al. Retinal vasculature fractal and stroke mortality[ J]. Stroke, 2021, 52(4): 1276-1282. DOI: 10.1161/ STROKEAHA.120.031886.
5、Cheung CY, Chan VTT, Mok VC, et al. Potential retinal biomarkers for dementia: what is new[ J]. Curr Opin Neurol, 2019, 32(1): 82-91. DOI: 10.1097/WCO.0000000000000645.Cheung CY, Chan VTT, Mok VC, et al. Potential retinal biomarkers for dementia: what is new[ J]. Curr Opin Neurol, 2019, 32(1): 82-91. DOI: 10.1097/WCO.0000000000000645.
6、Betzler BK, Rim TH, Sabanayagam C, et al. Artificial intelligence in predicting systemic parameters and diseases from ophthalmic imaging[ J]. Front Digit Health, 2022, 4: 889445. DOI: 10.3389/ fdgth.2022.889445.Betzler BK, Rim TH, Sabanayagam C, et al. Artificial intelligence in predicting systemic parameters and diseases from ophthalmic imaging[ J]. Front Digit Health, 2022, 4: 889445. DOI: 10.3389/ fdgth.2022.889445.
7、Khan NC, Perera C, Dow ER, et al. Predicting systemic health features from retinal fundus images using transfer-learning-based artificial intelligence models[ J]. Diagnostics, 2022, 12(7): 1714. DOI: 10.3390/ diagnostics12071714.Khan NC, Perera C, Dow ER, et al. Predicting systemic health features from retinal fundus images using transfer-learning-based artificial intelligence models[ J]. Diagnostics, 2022, 12(7): 1714. DOI: 10.3390/ diagnostics12071714.
8、Klein%20R%2C%20Sharrett%20AR%2C%20Klein%20BE%2C%20et%20al.%20Are%20retinal%20arteriolar%20abnormalities%20%0Arelated%20to%20atherosclerosis%3F%20%3A%20the%20Atherosclerosis%20Risk%20in%20Communities%20%0AStudy%5B%20J%5D.%20Arterioscler%20Thromb%20Vasc%20Biol%2C%202000%2C%2020(6)%3A%201644-1650.%20%0ADOI%3A%2010.1161%2F01.atv.20.6.1644.Klein%20R%2C%20Sharrett%20AR%2C%20Klein%20BE%2C%20et%20al.%20Are%20retinal%20arteriolar%20abnormalities%20%0Arelated%20to%20atherosclerosis%3F%20%3A%20the%20Atherosclerosis%20Risk%20in%20Communities%20%0AStudy%5B%20J%5D.%20Arterioscler%20Thromb%20Vasc%20Biol%2C%202000%2C%2020(6)%3A%201644-1650.%20%0ADOI%3A%2010.1161%2F01.atv.20.6.1644.
9、Ikram%20MK%2C%20de%20Jong%20FJ%2C%20Vingerling%20JR%2C%20et%20al.%20Are%20retinal%20arteriolar%20or%0Avenular%20diameters%20associated%20with%20markers%20for%20cardiovascular%20disorders%3F%20%0AThe%20Rotterdam%20Study%5B%20J%5D.%20Invest%20Ophthalmol%20Vis%20Sci%2C%202004%2C%2045(7)%3A%20%0A2129-2134.%20DOI%3A%2010.1167%2Fiovs.03-1390.Ikram%20MK%2C%20de%20Jong%20FJ%2C%20Vingerling%20JR%2C%20et%20al.%20Are%20retinal%20arteriolar%20or%0Avenular%20diameters%20associated%20with%20markers%20for%20cardiovascular%20disorders%3F%20%0AThe%20Rotterdam%20Study%5B%20J%5D.%20Invest%20Ophthalmol%20Vis%20Sci%2C%202004%2C%2045(7)%3A%20%0A2129-2134.%20DOI%3A%2010.1167%2Fiovs.03-1390.
10、Bakalli%20A%2C%20Ko%C3%A7inaj%20D%2C%20Bakalli%20A%2C%20et%20al.%20Relationship%20of%20hypertensive%20%0Aretinopathy%20to%20thoracic%20aortic%20atherosclerosis%20in%20patients%20with%20severe%20%0Aarterial%20hypertension%5B%20J%5D.%20Clin%20Exp%20Hypertens%2C%202011%2C%2033(2)%3A%2089-94.%20%0ADOI%3A%2010.3109%2F10641963.2010.503307.Bakalli%20A%2C%20Ko%C3%A7inaj%20D%2C%20Bakalli%20A%2C%20et%20al.%20Relationship%20of%20hypertensive%20%0Aretinopathy%20to%20thoracic%20aortic%20atherosclerosis%20in%20patients%20with%20severe%20%0Aarterial%20hypertension%5B%20J%5D.%20Clin%20Exp%20Hypertens%2C%202011%2C%2033(2)%3A%2089-94.%20%0ADOI%3A%2010.3109%2F10641963.2010.503307.
11、Cheung N, Bluemke DA , K lein R , et al. R etinal ar ter iolar narrowing and left ventricular remodeling: the multi-ethnic study of atherosclerosis[ J]. J Am Coll Cardiol, 2007, 50(1): 48-55. DOI: 10.1016/j.jacc.2007.03.029.Cheung N, Bluemke DA , K lein R , et al. R etinal ar ter iolar narrowing and left ventricular remodeling: the multi-ethnic study of atherosclerosis[ J]. J Am Coll Cardiol, 2007, 50(1): 48-55. DOI: 10.1016/j.jacc.2007.03.029.
12、Wang L, Wong TY, Sharrett AR, et al. Relationship between retinal arteriolar narrowing and myocardial perfusion: multi-ethnic study of atherosclerosis[ J]. Hypertension, 2008, 51(1): 119-126. DOI: 10.1161/HYPERTENSIONAHA.107.098343.Wang L, Wong TY, Sharrett AR, et al. Relationship between retinal arteriolar narrowing and myocardial perfusion: multi-ethnic study of atherosclerosis[ J]. Hypertension, 2008, 51(1): 119-126. DOI: 10.1161/HYPERTENSIONAHA.107.098343.
13、Fu Y, Yusufu M, Wang Y, et al. Association of retinal microvascular density and complexity with incident coronary heart disease[ J]. Atherosclerosis, 2023, 380: 117196. DOI: 10.1016/j.atherosclerosis. 2023.117196.Fu Y, Yusufu M, Wang Y, et al. Association of retinal microvascular density and complexity with incident coronary heart disease[ J]. Atherosclerosis, 2023, 380: 117196. DOI: 10.1016/j.atherosclerosis. 2023.117196.
14、Gerrits N, Elen B, Craenendonck TV, et al. Age and sex affect deep learning prediction of cardiometabolic risk factors from retinal images[ J]. Sci Rep, 2020, 10(1): 9432. DOI: 10.1038/s41598-020- 65794-4.Gerrits N, Elen B, Craenendonck TV, et al. Age and sex affect deep learning prediction of cardiometabolic risk factors from retinal images[ J]. Sci Rep, 2020, 10(1): 9432. DOI: 10.1038/s41598-020- 65794-4.
15、Kim YD, Noh KJ, Byun SJ, et al. Effects of hypertension, diabetes, and smoking on age and sex prediction from retinal fundus images[ J]. Sci Rep, 2020, 10(1): 4623. DOI: 10.1038/s41598-020-61519-9.Kim YD, Noh KJ, Byun SJ, et al. Effects of hypertension, diabetes, and smoking on age and sex prediction from retinal fundus images[ J]. Sci Rep, 2020, 10(1): 4623. DOI: 10.1038/s41598-020-61519-9.
16、Iao WC, Zhang W, Wang X, et al. Deep learning algorithms for screening and diagnosis of systemic diseases based on ophthalmic manifestations: a systematic review[ J]. Diagnostics, 2023, 13(5): 900. DOI: 10.3390/diagnostics13050900.Iao WC, Zhang W, Wang X, et al. Deep learning algorithms for screening and diagnosis of systemic diseases based on ophthalmic manifestations: a systematic review[ J]. Diagnostics, 2023, 13(5): 900. DOI: 10.3390/diagnostics13050900.
17、Shigueoka LS, Mariottoni EB, Thompson AC, et al. Predicting age from optical coherence tomography scans with deep learning[ J]. Transl Vis Sci Technol, 2021, 10(1): 12. DOI: 10.1167/tvst.10.1.12.Shigueoka LS, Mariottoni EB, Thompson AC, et al. Predicting age from optical coherence tomography scans with deep learning[ J]. Transl Vis Sci Technol, 2021, 10(1): 12. DOI: 10.1167/tvst.10.1.12.
18、Chueh KM, Hsieh YT, Chen HH, et al. Identification of sex and age from macular optical coherence tomography and feature analysis using deep learning[ J]. Am J Ophthalmol, 2022, 235: 221-228. DOI: 10.1016/j.ajo.2021.09.015.Chueh KM, Hsieh YT, Chen HH, et al. Identification of sex and age from macular optical coherence tomography and feature analysis using deep learning[ J]. Am J Ophthalmol, 2022, 235: 221-228. DOI: 10.1016/j.ajo.2021.09.015.
19、Rim TH, Lee G, Kim Y, et al. Prediction of systemic biomarkers from retinal photographs: development and validation of deep-learning algorithms[ J]. Lancet Digit Health, 2020, 2(10): e526-e536. DOI: 10.1016/S2589-7500(20)30216-8.Rim TH, Lee G, Kim Y, et al. Prediction of systemic biomarkers from retinal photographs: development and validation of deep-learning algorithms[ J]. Lancet Digit Health, 2020, 2(10): e526-e536. DOI: 10.1016/S2589-7500(20)30216-8.
20、Poplin R, Varadarajan AV, Blumer K, et al. Prediction of cardiovascular risk factors from retinal fundus photographs via deep learning[ J]. Nat Biomed Eng, 2018, 2(3): 158-164. DOI: 10.1038/s41551-018-0195-0.Poplin R, Varadarajan AV, Blumer K, et al. Prediction of cardiovascular risk factors from retinal fundus photographs via deep learning[ J]. Nat Biomed Eng, 2018, 2(3): 158-164. DOI: 10.1038/s41551-018-0195-0.
21、Wilson PW, D'Agostino RB, Levy D, et al. Prediction of coronary heart disease using risk factor categories[ J]. Circulation, 1998, 97(18): 1837-1847. DOI: 10.1161/01.cir.97.18.1837.Wilson PW, D'Agostino RB, Levy D, et al. Prediction of coronary heart disease using risk factor categories[ J]. Circulation, 1998, 97(18): 1837-1847. DOI: 10.1161/01.cir.97.18.1837.
22、Stone NJ, Robinson JG, Lichtenstein AH, et al. 2013 ACC/ AHA guideline on the treatment of blood cholesterol to reduce atherosclerotic cardiovascular risk in adults: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines[ J]. J Am Coll Cardiol, 2014, 63(25 Pt B): 2889- 2934. DOI: 10.1016/j.jacc.2013.11.002.Stone NJ, Robinson JG, Lichtenstein AH, et al. 2013 ACC/ AHA guideline on the treatment of blood cholesterol to reduce atherosclerotic cardiovascular risk in adults: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines[ J]. J Am Coll Cardiol, 2014, 63(25 Pt B): 2889- 2934. DOI: 10.1016/j.jacc.2013.11.002.
23、Hira RS, Kennedy K, Nambi V, et al. Frequency and practice-level variation in inappropriate aspirin use for the primary prevention of cardiovascular disease: insights from the National Cardiovascular Disease Registry 's Practice Innovation and Clinical Excellence registry[ J]. J Am Coll Cardiol, 2015, 65(2): 111-121. DOI: 10.1016/ j.jacc.2014.10.035.Hira RS, Kennedy K, Nambi V, et al. Frequency and practice-level variation in inappropriate aspirin use for the primary prevention of cardiovascular disease: insights from the National Cardiovascular Disease Registry 's Practice Innovation and Clinical Excellence registry[ J]. J Am Coll Cardiol, 2015, 65(2): 111-121. DOI: 10.1016/ j.jacc.2014.10.035.
24、Yi JK, Rim TH, Park S, et al. Cardiovascular disease risk assessment using a deep-learning-based retinal biomarker: a comparison with existing risk scores[ J]. Eur Heart J Digit Health, 2023, 4(3): 236-244. DOI: 10.1093/ehjdh/ztad023.Yi JK, Rim TH, Park S, et al. Cardiovascular disease risk assessment using a deep-learning-based retinal biomarker: a comparison with existing risk scores[ J]. Eur Heart J Digit Health, 2023, 4(3): 236-244. DOI: 10.1093/ehjdh/ztad023.
25、Xu J, Xue K, Zhang K. Current status and future trends of clinical diagnoses via image-based deep learning[ J]. Theranostics, 2019, 9(25): 7556-7565. DOI: 10.7150/thno.38065.Xu J, Xue K, Zhang K. Current status and future trends of clinical diagnoses via image-based deep learning[ J]. Theranostics, 2019, 9(25): 7556-7565. DOI: 10.7150/thno.38065.
26、Zhu Z, Chen Y, Wang W, et al. Association of retinal age gap with arterial stiffness and incident cardiovascular disease[ J]. Stroke, 2022, 53(11): 3320-3328. DOI: 10.1161/STROKEAHA.122.038809.Zhu Z, Chen Y, Wang W, et al. Association of retinal age gap with arterial stiffness and incident cardiovascular disease[ J]. Stroke, 2022, 53(11): 3320-3328. DOI: 10.1161/STROKEAHA.122.038809.
27、Chen C, Bickford ME, Hirsch JA. Untangling the web between eye and brain[ J]. Cell, 2016, 165(1): 20-21. DOI: 10.1016/j.cell.2016.03.010.Chen C, Bickford ME, Hirsch JA. Untangling the web between eye and brain[ J]. Cell, 2016, 165(1): 20-21. DOI: 10.1016/j.cell.2016.03.010.
28、Zhang Q, Li J, Bian M, et al. Retinal imaging techniques based on machine learning models in recognition and prediction of mild cognitive impairment[ J]. Neuropsychiatr Dis Treat, 2021, 17: 3267- 3281. DOI: 10.2147/NDT.S333833.Zhang Q, Li J, Bian M, et al. Retinal imaging techniques based on machine learning models in recognition and prediction of mild cognitive impairment[ J]. Neuropsychiatr Dis Treat, 2021, 17: 3267- 3281. DOI: 10.2147/NDT.S333833.
29、Tian J, Smith G, Guo H, et al. Modular machine learning for Alzheimer's disease classification from retinal vasculature[ J]. Sci Rep, 2021, 11(1): 238. DOI: 10.1038/s41598-020-80312-2.Tian J, Smith G, Guo H, et al. Modular machine learning for Alzheimer's disease classification from retinal vasculature[ J]. Sci Rep, 2021, 11(1): 238. DOI: 10.1038/s41598-020-80312-2.
30、Cheung CY, Ran AR, Wang S, et al. A deep learning model for detection of Alzheimer's disease based on retinal photographs: a retrospective, multicentre case-control study[ J]. Lancet Digit Health, 2022, 4(11): e806-e815. DOI: 10.1016/S2589-7500(22)00169-8.Cheung CY, Ran AR, Wang S, et al. A deep learning model for detection of Alzheimer's disease based on retinal photographs: a retrospective, multicentre case-control study[ J]. Lancet Digit Health, 2022, 4(11): e806-e815. DOI: 10.1016/S2589-7500(22)00169-8.
31、Wisely CE, Wang D, Henao R, et al. Convolutional neural network to identify symptomatic Alzheimer's disease using multimodal retinal imaging[ J]. Br J Ophthalmol, 2022, 106(3): 388-395. DOI: 10.1136/ bjophthalmol-2020-317659.Wisely CE, Wang D, Henao R, et al. Convolutional neural network to identify symptomatic Alzheimer's disease using multimodal retinal imaging[ J]. Br J Ophthalmol, 2022, 106(3): 388-395. DOI: 10.1136/ bjophthalmol-2020-317659.
32、Arevalo-Rodriguez I, Smailagic N, Roqué-Figuls M, et al. Mini-Mental State Examination (MMSE) for the early detection of dementia in people with mild cognitive impairment (MCI)[ J]. Cochrane Database Syst Rev, 2021, 7(7): CD010783. DOI: 10.1002/14651858.CD010783. pub3.Arevalo-Rodriguez I, Smailagic N, Roqué-Figuls M, et al. Mini-Mental State Examination (MMSE) for the early detection of dementia in people with mild cognitive impairment (MCI)[ J]. Cochrane Database Syst Rev, 2021, 7(7): CD010783. DOI: 10.1002/14651858.CD010783. pub3.
33、Kim J, Jeong M, Stiles WR , et al. Neuroimaging modalities in Alzheimer's disease: diagnosis and clinical features[ J]. Int J Mol Sci, 2022, 23(11): 6079. DOI: 10.3390/ijms23116079.Kim J, Jeong M, Stiles WR , et al. Neuroimaging modalities in Alzheimer's disease: diagnosis and clinical features[ J]. Int J Mol Sci, 2022, 23(11): 6079. DOI: 10.3390/ijms23116079.
34、Hui HYH, Ran AR, Dai JJ, et al. Deep reinforcement learning-based retinal imaging in Alzheimer's disease: potential and perspectives[ J]. J Alzheimers Dis, 2023, 94(1): 39-50. DOI: 10.3233/JAD-230055.Hui HYH, Ran AR, Dai JJ, et al. Deep reinforcement learning-based retinal imaging in Alzheimer's disease: potential and perspectives[ J]. J Alzheimers Dis, 2023, 94(1): 39-50. DOI: 10.3233/JAD-230055.
35、Nunes A, Silva G, Duque C, et al. Retinal texture biomarkers may help to discriminate between Alzheimer's, Parkinson's, and healthy controls[ J]. PLoS One, 2019, 14(6): e0218826. DOI: 10.1371/journal. pone.0218826.Nunes A, Silva G, Duque C, et al. Retinal texture biomarkers may help to discriminate between Alzheimer's, Parkinson's, and healthy controls[ J]. PLoS One, 2019, 14(6): e0218826. DOI: 10.1371/journal. pone.0218826.
36、Reiner J, Franken L, Raveh E, et al. Oculometric measures as a tool for assessment of clinical symptoms and severity of Parkinson's disease[ J]. J Neural Transm, 2023, 130(10): 1241-1248. DOI: 10.1007/s00702- 023-02681-y.Reiner J, Franken L, Raveh E, et al. Oculometric measures as a tool for assessment of clinical symptoms and severity of Parkinson's disease[ J]. J Neural Transm, 2023, 130(10): 1241-1248. DOI: 10.1007/s00702- 023-02681-y.
37、Ahn S, Shin J, Song SJ, et al. Neurologic dysfunction assessment in parkinson disease based on fundus photographs using deep learning[ J]. JAMA Ophthalmol, 2023, 141(3): 234-240. DOI: 10.1001/ jamaophthalmol.2022.5928.Ahn S, Shin J, Song SJ, et al. Neurologic dysfunction assessment in parkinson disease based on fundus photographs using deep learning[ J]. JAMA Ophthalmol, 2023, 141(3): 234-240. DOI: 10.1001/ jamaophthalmol.2022.5928.
38、Hu W, Wang W, Wang Y, et al. Retinal age gap as a predictive biomarker of future risk of Parkinson's disease[ J]. Age Ageing, 2022, 51(3): afac062. DOI: 10.1093/ageing/afac062.Hu W, Wang W, Wang Y, et al. Retinal age gap as a predictive biomarker of future risk of Parkinson's disease[ J]. Age Ageing, 2022, 51(3): afac062. DOI: 10.1093/ageing/afac062.
39、Montol%C3%ADo%20A%2C%20Mart%C3%ADn-Gallego%20A%2C%20Cego%C3%B1ino%20J%2C%20et%20al.%20Machine%20learning%20in%20%0Adiagnosis%20and%20disability%20prediction%20of%20multiple%20sclerosis%20using%20optical%20%0Acoherence%20tomography%5B%20J%5D.%20Comput%20Biol%20Med%2C%202021%2C%20133%3A%20104416.%20%0ADOI%3A%2010.1016%2Fj.compbiomed.2021.104416.Montol%C3%ADo%20A%2C%20Mart%C3%ADn-Gallego%20A%2C%20Cego%C3%B1ino%20J%2C%20et%20al.%20Machine%20learning%20in%20%0Adiagnosis%20and%20disability%20prediction%20of%20multiple%20sclerosis%20using%20optical%20%0Acoherence%20tomography%5B%20J%5D.%20Comput%20Biol%20Med%2C%202021%2C%20133%3A%20104416.%20%0ADOI%3A%2010.1016%2Fj.compbiomed.2021.104416.
40、López-Dorado A, Ortiz M, Satue M, et al. Early diagnosis of multiple sclerosis using swept-source optical coherence tomography and convolutional neural networks trained with data augmentation[ J]. Sensors, 2021, 22(1): 167. DOI: 10.3390/s22010167.López-Dorado A, Ortiz M, Satue M, et al. Early diagnosis of multiple sclerosis using swept-source optical coherence tomography and convolutional neural networks trained with data augmentation[ J]. Sensors, 2021, 22(1): 167. DOI: 10.3390/s22010167.
41、Cavaliere C, Vilades E, Alonso-Rodríguez MC, et al. Computer-aided diagnosis of multiple sclerosis using a support vector machine and optical coherence tomography features[ J]. Sensors, 2019, 19(23): 5323. DOI: 10.3390/s19235323.Cavaliere C, Vilades E, Alonso-Rodríguez MC, et al. Computer-aided diagnosis of multiple sclerosis using a support vector machine and optical coherence tomography features[ J]. Sensors, 2019, 19(23): 5323. DOI: 10.3390/s19235323.
42、Bodaghi B, Massamba N, Izzedine H. The eye: a window on kidney diseases[ J]. Clin Kidney J, 2014, 7(4): 337-338. DOI: 10.1093/ckj/ sfu073.Bodaghi B, Massamba N, Izzedine H. The eye: a window on kidney diseases[ J]. Clin Kidney J, 2014, 7(4): 337-338. DOI: 10.1093/ckj/ sfu073.
43、Booij JC, Baas DC, Beisekeeva J, et al. The dynamic nature of Bruch's membrane[ J]. Prog Retin Eye Res, 2010, 29(1): 1-18. DOI: 10.1016/ j.preteyeres.2009.08.003.Booij JC, Baas DC, Beisekeeva J, et al. The dynamic nature of Bruch's membrane[ J]. Prog Retin Eye Res, 2010, 29(1): 1-18. DOI: 10.1016/ j.preteyeres.2009.08.003.
44、Grunwald JE, Alexander J, Ying GS, et al. Retinopathy and chronic kidney disease in the Chronic Renal Insufficiency Cohort (CRIC) study[ J]. Arch Ophthalmol, 2012, 130(9): 1136-1144. DOI: 10.1001/ archophthalmol.2012.1800.Grunwald JE, Alexander J, Ying GS, et al. Retinopathy and chronic kidney disease in the Chronic Renal Insufficiency Cohort (CRIC) study[ J]. Arch Ophthalmol, 2012, 130(9): 1136-1144. DOI: 10.1001/ archophthalmol.2012.1800.
45、Weiner DE, Tighiouart H, Reynolds R , et al. Kidney function, albuminuria and age-related macular degeneration in NHANES III[ J]. Nephrol Dial Transplant, 2011, 26(10): 3159-3165. DOI: 10.1093/ ndt/gfr022.Weiner DE, Tighiouart H, Reynolds R , et al. Kidney function, albuminuria and age-related macular degeneration in NHANES III[ J]. Nephrol Dial Transplant, 2011, 26(10): 3159-3165. DOI: 10.1093/ ndt/gfr022.
46、Wong CW, Lamoureux EL, Cheng CY, et al. Increased burden of vision impairment and eye diseases in persons with chronic kidney disease - A population-based study[ J]. EBioMedicine, 2016, 5: 193-197. DOI: 10.1016/j.ebiom.2016.01.023.Wong CW, Lamoureux EL, Cheng CY, et al. Increased burden of vision impairment and eye diseases in persons with chronic kidney disease - A population-based study[ J]. EBioMedicine, 2016, 5: 193-197. DOI: 10.1016/j.ebiom.2016.01.023.
47、Liew G, Mitchell P, Wong TY, et al. Retinal microvascular signs are associated with chronic kidney disease in persons with and without diabetes[ J]. Kidney Blood Press Res, 2012, 35(6): 589-594. DOI: 10.1159/000339173.Liew G, Mitchell P, Wong TY, et al. Retinal microvascular signs are associated with chronic kidney disease in persons with and without diabetes[ J]. Kidney Blood Press Res, 2012, 35(6): 589-594. DOI: 10.1159/000339173.
48、Kang EYC, Hsieh YT, Li CH, et al. Deep learning-based detection of early renal function impairment using retinal fundus images: model development and validation[ J]. JMIR Med Inform, 2020, 8(11): e23472. DOI: 10.2196/23472.Kang EYC, Hsieh YT, Li CH, et al. Deep learning-based detection of early renal function impairment using retinal fundus images: model development and validation[ J]. JMIR Med Inform, 2020, 8(11): e23472. DOI: 10.2196/23472.
49、Betzler BK, Chee EYL, He F, et al. Deep learning algorithms to detect diabetic kidney disease from retinal photographs in multiethnic populations with diabetes[ J]. J Am Med Inform Assoc, 2023, 30(12): 1904-1914. DOI: 10.1093/jamia/ocad179.Betzler BK, Chee EYL, He F, et al. Deep learning algorithms to detect diabetic kidney disease from retinal photographs in multiethnic populations with diabetes[ J]. J Am Med Inform Assoc, 2023, 30(12): 1904-1914. DOI: 10.1093/jamia/ocad179.
50、Sabanayagam C, Xu D, Ting DSW, et al. A deep learning algorithm to detect chronic kidney disease from retinal photographs in communitybased populations[ J]. Lancet Digit Health, 2020, 2(6): e295-e302. DOI: 10.1016/S2589-7500(20)30063-7.Sabanayagam C, Xu D, Ting DSW, et al. A deep learning algorithm to detect chronic kidney disease from retinal photographs in communitybased populations[ J]. Lancet Digit Health, 2020, 2(6): e295-e302. DOI: 10.1016/S2589-7500(20)30063-7.
51、Zhang S, Chen R, Wang Y, et al. Association of retinal age gap and risk of kidney failure: a UK biobank study[ J]. Am J Kidney Dis, 2023, 81(5): 537-544.e1. DOI: 10.1053/j.ajkd.2022.09.018.Zhang S, Chen R, Wang Y, et al. Association of retinal age gap and risk of kidney failure: a UK biobank study[ J]. Am J Kidney Dis, 2023, 81(5): 537-544.e1. DOI: 10.1053/j.ajkd.2022.09.018.
52、Liu Y, Zhang F, Gao X, et al. Lesion-aware attention network for diabetic nephropathy diagnosis with optical coherence tomography images[ J]. Front Med, 2023, 10: 1259478. DOI: 10.3389/ fmed.2023.1259478.Liu Y, Zhang F, Gao X, et al. Lesion-aware attention network for diabetic nephropathy diagnosis with optical coherence tomography images[ J]. Front Med, 2023, 10: 1259478. DOI: 10.3389/ fmed.2023.1259478.
53、Bobrov E, Georgievskaya A, Kiselev K, et al. PhotoAgeClock: deep learning algorithms for development of non-invasive visual biomarkers of aging[ J]. Aging, 2018, 10(11): 3249-3259. DOI: 10.18632/ aging.101629.Bobrov E, Georgievskaya A, Kiselev K, et al. PhotoAgeClock: deep learning algorithms for development of non-invasive visual biomarkers of aging[ J]. Aging, 2018, 10(11): 3249-3259. DOI: 10.18632/ aging.101629.
54、Ma J, Xu X, Li M, et al. Predictive models of aging of the human eye based on ocular anterior segment morphology[ J]. J Biomed Inform, 2021, 120: 103855. DOI: 10.1016/j.jbi.2021.103855.Ma J, Xu X, Li M, et al. Predictive models of aging of the human eye based on ocular anterior segment morphology[ J]. J Biomed Inform, 2021, 120: 103855. DOI: 10.1016/j.jbi.2021.103855.
55、Nusinovici S, Rim TH, Yu M, et al. Retinal photograph-based deep learning predicts biological age, and stratifies morbidity and mortality risk[ J]. Age Ageing, 2022, 51(4): afac065. DOI: 10.1093/ageing/ afac065.Nusinovici S, Rim TH, Yu M, et al. Retinal photograph-based deep learning predicts biological age, and stratifies morbidity and mortality risk[ J]. Age Ageing, 2022, 51(4): afac065. DOI: 10.1093/ageing/ afac065.
56、Zhu Z, Shi D, Peng G, et al. Retinal age gap as a predictive biomarker for mortality risk[ J]. Br J Ophthalmol, 2023, 107(4): 547-554. DOI: 10.1136/bjophthalmol-2021-319807.Zhu Z, Shi D, Peng G, et al. Retinal age gap as a predictive biomarker for mortality risk[ J]. Br J Ophthalmol, 2023, 107(4): 547-554. DOI: 10.1136/bjophthalmol-2021-319807.
57、Li R, Chen W, Li M, et al. LensAge index as a deep learning-based biological age for self-monitoring the risks of age-related diseases and mortality[ J]. Nat Commun, 2023, 14(1): 7126. DOI: 10.1038/s41467- 023-42934-8.Li R, Chen W, Li M, et al. LensAge index as a deep learning-based biological age for self-monitoring the risks of age-related diseases and mortality[ J]. Nat Commun, 2023, 14(1): 7126. DOI: 10.1038/s41467- 023-42934-8.
上一篇
下一篇
其他期刊
  • 眼科学报

    主管:中华人民共和国教育部
    主办:中山大学
    承办:中山大学中山眼科中心
    主编:林浩添
    主管:中华人民共和国教育部
    主办:中山大学
    浏览
  • Eye Science

    主管:中华人民共和国教育部
    主办:中山大学
    承办:中山大学中山眼科中心
    主编:林浩添
    主管:中华人民共和国教育部
    主办:中山大学
    浏览
推荐阅读
出版者信息
目录