您的位置: 首页 > 2025年1月 第40卷 第1期 > 文字全文
2023年7月 第38卷 第7期11
目录

人工智能在泪器疾病诊疗中的应用:挑战与机遇

Application of artificial intelligence in the diagnosis and treatment of lacrimal disorders: challenges and opportunities

来源期刊: 眼科学报 | 2025年1月 第40卷 第1期 53-66 发布时间:2025-1-28 收稿时间:2025/1/17 9:34:04 阅读量:110
作者:
关键词:
泪器疾病人工智能机器学习精准医疗
lacrimal disorders artificial intelligence machine learning precision medicine
DOI:
10.12419/24093010
收稿时间:
2024-09-30 
修订日期:
2024-12-10 
接收日期:
2024-12-30 
泪器疾病是一类常见的眼科疾病,其诊疗过程复杂,治疗方法精细,涉及多种临床数据及影像资料。现有研究表明,随着人工智能(artificial intelligence,AI)技术,尤其是机器学习和深度学习的发展,AI在泪器疾病的早期筛查、精确诊断和个性化治疗中展现了巨大的应用潜力。AI能够通过高效的图像分析、多模态数据融合及深度学习算法,提供更加精确的疾病识别和治疗方案,并且能够对患者的病情进行定期监测和动态调整,提升治疗效果。然而,其仍面临诸多挑战,如多模态数据融合的复杂性、模型泛化能力的局限以及实时预测和动态调整的需求等,需要通过持续的技术创新、算法优化和跨学科合作来实现。文章对当前AI在泪器疾病诊疗中的应用现状进行了全面梳理和总结,深入分析了AI技术在诊断与治疗中的优势与局限,特别强调了AI与新兴技术的结合在优化临床决策支持系统方面的重要性。通过分析现有的挑战与技术融合策略,文章提出了AI在泪器疾病诊疗中的发展方向,旨在为未来的研究者提供创新性的思路,为眼科领域的临床实践提供有价值的参考,助力泪器疾病的精准医疗和个性化治疗的发展。
Lacrimal disorders are common ophthalmic conditions characterized by complex diagnostic and treatment processes, involving intricate therapeutic approaches and diverse clinical and imaging data. Recent studies have indicated that with the advancements in artificial intelligence (AI) technologies, particularly in machine learning and deep learning, AI demonstrates significant potential in the early screening, accurate diagnosis, and personalized treatment of lacrimal disorders. AI has the ability to provide more precise disease identification and treatment strategies through efficient image analysis, multimodal data fusion, and deep learning algorithms. Additionally, it enables regular monitoring and dynamic adjustment of patients' conditions, improving treatment outcomes. However, several challenges persist, such as the complexity of multimodal data integration, limitations in model generalization capabilities, and the need for real-time prediction and dynamic adjustments, all of which necessitate continuous technological innovations, algorithm optimization, and interdisciplinary collaborations. This paper provides a comprehensive review of the current status of AI applications in the diagnosis and treatment of lacrimal disorders, analyzing the advantages and limitations of AI in clinical practice. It especially emphasizes the importance of integrating AI with emerging technologies to optimize clinical decision support systems. By addressing the existing challenges and exploring strategies for technological integration, this paper proposes future directions for the development of AI in lacrimal disorder diagnosis and treatment, aiming to offer innovative perspectives for future researchers and valuable references for clinical practice in the field of ophthalmology, ultimately contributing to the advancement of precision medicine and personalized treatment for lacrimal disorders.

文章亮点

1. 关键发现

 • 通过总结 AI 在泪器疾病诊疗的国内外研究现状,发现 AI 技术能有效提高泪器疾病的诊断准确性,特别是在疾病的早期筛查和治疗方案制定中展现出显著优势。

2. 已知与发现

 • 随着深度学习和机器学习的发展,AI 技术已应用于泪器疾病的智能化影像分析。
 • AI 辅助的泪器疾病诊断系统能够通过分析眼表指标提供早期诊断依据。
 • 虚拟现实技术结合混合现实技术辅助泪器疾病的手术治疗。
 • 结合多维度生物标志物的智能化泪液成分分析系统,可无创、快速地评估泪液质量。

3. 意义与改变

 • 意义与改变将 AI 技术与泪器疾病诊疗深度融合,能够显著提高诊断效率、缩短诊疗周期、制定个性化治疗方案及辅助预后预测,为泪器疾病的早期干预和精准医疗开辟了新的路径。

       泪器疾病作为眼科常见且较为复杂的疾病,涵盖了泪道阻塞、炎症、占位性病变等多种病理状态,其发病机制复杂,临床表现多样。其诊疗过程需要大量的临床数据支持,包括患者的病史、影像学检查结果以及实验室检测数据,且治疗方案往往需根据患者的病情进行个性化调整。因此,泪器疾病的治疗既要求医生具备丰富的临床经验,又需要依靠精细化的诊断手段和智能化的治疗策略。
       传统的泪器疾病诊疗依赖于临床表现、体格检查、泪液分析等常规手段,但此类方法存在一定的局限性。随着人工智能(artificial intelligence,AI)技术,特别是机器学习和深度学习的发展,AI在医学影像分析、数据处理和临床决策支持中展现出了巨大的潜力。通过对大数据的深度学习和模式识别,AI能够在泪器疾病的早期筛查、病因分析、诊断准确性和个性化治疗方案制定等方面提供全新的思路和方法。AI在泪器疾病的应用,不仅能够提高诊断效率,缩短诊疗周期,同时能够实现多维数据的智能融合和精准预测,为临床医生提供更加科学的决策依据。
       尽管AI在泪器疾病的诊疗中展现了显著的优势,但其仍面临一系列挑战,主要包括多模态数据的融合、AI模型泛化能力的局限、实时动态预测和治疗方案调整的需求等问题。因此,本文旨在综述当前AI在泪器疾病诊疗中的应用现状,分析其面临的技术难题,探讨新兴技术与AI的融合方向,并提出未来发展的创新性思路。通过前瞻性的分析与讨论,本文旨在为眼科领域的研究者和临床实践提供有价值的参考,推动AI技术在泪器疾病诊疗中的进一步应用和发展。

1人工智能的基本概念及其在眼科领域的应用

       人工智能(artificial intelligence,AI)是计算机科学的一个重要分支,旨在使计算机具备类似于人类的学习、感知、理解、推理、归纳、决策和解决问题的能力[1-2]。其中,机器学习(machine learning,ML)和深度学习(deep learning,DL)的广泛应用,显著促进了医学诊疗领域的发展[3–5]。在眼科领域,AI最初应用于眼底照片和光学相干断层扫描(optical Coherence Tomography,OCT)图像的分析,这一早期应用主要集中在眼后段疾病的诊断[6-7]。随着DL算法的发展,AI技术现已扩展至眼前段结构的成像数据分析,包括前段照片、前段OCT(anterior segment optical Coherence Tomography,AS-OCT)图像、镜面显微镜图像、角膜地形图、体内共聚焦显微镜(In vivo confocal microscopy,IVCM)图像、红外睑板腺成像(meibography)以及泪液干涉测量等[8]。这些技术使得AI在眼前段疾病(角膜疾病、虹膜疾病以及泪器疾病等)的诊疗中显示出广阔的应用前景[9]

2泪器疾病概述

2.1传统诊疗方法的局限性

       泪器疾病作为眼科的一类常见疾病,涵盖了影响泪液分泌、流通和排出的多种病理状态[10-11]。其传统诊疗方法包括泪液分泌测试、泪道造影、药物治疗和手术治疗等[12–15]。尽管这些传统治疗方法在临床上已有一定的应用历史,但随着医学技术的发展和患者需求的提高,其在治疗复杂泪器疾病时的局限性逐渐显现,如主观性强、个体差异大、耐药性增强、检测能力有限、治疗效果不确定性高及潜在的并发症风险等[16–18]。此外,传统方法在数据收集和分析方面缺乏系统化,难以实现个性化治疗。因此,急需引入更先进的技术手段,以提升泪器疾病的诊疗效果,AI的应用或将为此带来新的机遇和变革,提供更为精准、个性化的诊疗方案。

2.2 AI在泪器疾病中的应用现状及前景

       在泪器疾病的诊疗过程中,AI技术正逐步发挥其独特的优势,改变传统的临床诊疗模式,图1所示为AI辅助泪器疾病诊疗的基本流程。首先,AI通过收集患者的多模态数据,包括医学影像、电子病历和临床数据,进行预处理和特征提取。随后,利用DL技术对影像数据和文本数据进行分析,提取关键特征,并实现多模态数据融合,以增强特征的表达。在此基础上,AI算法对融合特征进行深入分析,为泪器疾病的诊断和治疗提供科学依据。进一步,其将分析结果整合至临床决策支持系统,辅助医生制定个性化治疗方案。同时,通过实时监测患者状态,动态调整治疗策略,以提高治疗效果。最后,AI进行临床验证与优化环节,确保模型的准确性和实用性,不仅提高了诊疗效率,也为泪器疾病的早期筛查、精准诊断和个性化治疗提供了新的思路。相关研究表明,AI在泪器疾病的筛查和诊断中取得了显著进展。Imamura等[19]利用DL技术,开发了一种基于前段光学相干断层成像(Anterior Segment Optical Coherence Tomography,AS-OCT)图像的自动筛查系统,能够高效识别泪道阻塞患者的泪道情况。该系统通过DL算法对大量图像进行分析,提升了泪器疾病的识别准确性,并在临床应用中显著缩短了诊疗周期。与此同时,Song等[20]提出了一种非侵入性的ML筛查模型,针对泪囊炎的诊断建立了以眼表指标为基础的评估系统。研究表明,该模型能够有效地对泪河高度(tear meniscus height,TMH)、客观散点指数(objective scattering index,OSI)及其衍生指标等多维度数据进行分析。在分析过程中,OSI的整体变化和趋势可以作为自变量输入相关的统计分析或结果预测模型,而不是一个孤立的模型,大大提高了核心指标的数据丰度。进一步地,它直接反映了泪液在眼表的积累情况,显示泪囊炎(溢泪)的临床体征,从而为早期诊断和治疗提供了重要依据。这种基于ML的技术不仅提高了泪器疾病的识别率,还为早期干预提供了新的可能性。
图 1 AI辅助泪器疾病诊疗基本流程
Figure 1 Basic workflow of AI-assisted diagnosis and treatment of lacrimal diseases

       上述研究表明,AI在泪器疾病的诊疗中快速发展。未来,随着算法的不断优化、技术迁移应用以及数据积累的丰富,AI有望在泪器疾病的早期筛查、精准诊断和个性化治疗中发挥更大的作用,推动泪器疾病诊疗的发展与进步。

3 AI在泪器疾病诊断中的作用

3.1智能化影像分析

       近年来,ML常被用于医学影像分析,特别是应用于分割或标记任务。作为神经网络的一个子类型,卷积神经网络(convolutional neural network,CNN)专注于处理2D横截面图像,通常用于与MRI成像相关的各种任务,包括去噪、特定类型scans的超分辨率分析以及用于数据增强的图像合成[21-24]。现有研究表明,通过基于AI技术的医学图像分割,可以实现眼眶解剖结构的自动识别和标记[25-26]。Hamwood等[27]基于DL架构,通过串联的两个CNN,实现了在眼眶计算机断层扫描(computed tomography,CT)/磁共振成像(magnetic resonance imaging,MRI)图像中对眼眶骨骼结构的自动分割。Fu等[28]基于感知卷积神经网络(context-aware convolutional neural network,CA-CNN),通过AI技术自动进行CT影像中眼眶脓肿的分割和量化,为眼眶脓肿患者提供更加精准、高效的诊断方法。
       尽管目前针对泪器疾病的AI技术仍在探索阶段,但其在眼科其他疾病的成功应用为泪器疾病的AI研究提供了宝贵的经验。借助AI的迁移学习,可将眼眶CT影像分析算法迁移至泪器的影像分析中。例如,CNN在智能化分析眼部影像方面表现出色,其能够识别微小的结构变化并辅助医生进行准确诊断[29]。Kim等[30]采用DL算法,运用CNN开发了2个分类模型:其一用于识别阻塞的存在,其二用于确定阻塞位置。研究结果显示,该模型在检测是否存在泪道阻塞方面的准确率为99.3%,在定位方面的准确率为95.5%。该研究结论表明,集成DL模型可以显著提高泪道阻塞的诊断准确性,减少人为误差,提高诊断的精确度。此外,AI能够自动检测泪道中的病变(如泪道狭窄、阻塞和炎症等)并进行分类,提供快速准确的诊断结果,辅助医生制定治疗决策[19]。Han等[31]提出了一种基于深度堆叠网络(deep stacking network,DSN)的可解释泪囊炎诊断模型,研究者通过使用梯度增强分类器(bradient boosting classifier,GBC)对堆叠模块进行训练,并结合局部和全局解释方法,解决了模型的“黑箱”问题。采用定量眼部指标和决策曲线分析对受试对象进行临床效用评估。结果表明,该模型在分类性能上优于8个传统分类器和其他DL模型,能够有效监测和诊断泪囊炎,具有较高的准确性和可靠性,该研究为泪囊炎的智能化诊断提供了新的思路和方法。通过图像增强和去噪技术,AI还能够提升影像的质量,使泪器中的细微结构异常更易于检测。进一步地,AI系统可以自动进行泪道影像中的测量(如泪道直径、长度等)并进行定量分析,这些定量数据对评估病情的严重程度和监测治疗效果至关重要。

3.2智能化泪液成分分析

       正常的泪液量在维持眼表生理和眼舒适性中发挥着至关重要的作用。在泪器疾病的诊断中,自动化诊断方法已被广泛应用,如OCT能够自动分析泪道冲洗的数据,精准识别和分类泪道阻塞或其他异常,简化了操作流程[19,32–34]。Schirmer试验通过测量特定时间内泪液的分泌量来评估泪腺的功能,用于检测与泪液分泌有关的眼病,为临床诊断提供支持[34-35],但其主要测量泪腺的分泌能力,无法评估泪膜的稳定性、泪液的质量和眼表的整体健康状况。此外,传统的泪液分析依赖于显微镜和实验室技术,人工分析过程繁琐、误差大且耗时长。近年来,随着AI技术的发展,泪器疾病的辅助诊断逐渐由自动化向智能化转型。AI可以通过分析泪液的成分和含量,如泪液中的蛋白质、微量元素或其他生物标志物,提供对泪器功能和疾病状态的深入评估。Wang等[36]利用多通道卷积递归神经网络(convolutional neural network - gated recurrent unit,CNN-GRU)研发了由柔性微流控表皮贴片和基于DL神经网络的云服务器数据分析系统组成的智能化泪液成分检测系统。前者可收集泪液促进比色反应,识别维生素C、pH值、Ca²⁺和蛋白质等生物标志物并检测其含量,后者则嵌入到智能手机中,以实现颜色数据的采集、解释、智能化校正和显示。这种无创、快速、准确和同步监测泪液中多种关键生物标志物的方法,为评估眼部和全身健康状况提供了一种前景广阔、方便快捷的策略,为泪器疾病的诊断和精准医疗领域的发展提供了新思路。

3.3 智能化泪膜破裂时间预测

       干眼症(dry eye disease, DED)是最常见的眼科疾病之一,与泪器疾病密切相关。眼表泪液过度蒸发与泪腺功能减退或损伤导致的泪液分泌不足均会引发DED[37–39]。同时,DED的慢性表现可能促使泪器疾病的发展,且两者的症状可能重叠。在临床诊断中,医生需综合评估泪器健康状况,以助发现潜在的泪器疾病。目前,干眼症状与体征之间缺乏显著相关性,使得DED的诊断和监测面临较大挑战[40–42],DED的病理生理异质性及个体对干眼症状的感知差异是导致这种现象的潜在原因[43]。此外,现有的DED诊断测试的可靠性有限。传统干眼测试,如荧光素泪膜破裂时间(tear film break-up time,TFBUT),往往表现出较差的可靠性和可重复性[44]。为应对这些挑战,近年来,AI逐渐被引入到DED的诊断与监测中。Shimizu等[45]利用一种可录制并收集眼前段视频的裂隙灯设备,结合AI辅助制作了一种智能眼摄像头,基于ML模型验证每个案例的TFBUT数据集,使用TFBUT验证结果计算DED诊断性能[44,46-47],结果表明,这种使用眼表视频和DED诊断标准创建的DED诊断AI算法在预估泪膜破裂时间方面具有较高的准确性,表现出良好的应用前景。

表 1 AI在泪器疾病诊疗中的应用
Table 1 Applications of ai in the diagnosis and treatment of lacrimal diseases

第一作者

研究目的

数据集

AI算法

 

Imamura H[19]

智能化筛查泪道阻塞

AS-OCT图像数据集

CNN(ResNet-50)

Song X[20]

非侵入性筛查泪囊炎

眼表指标数据集

(泪膜破裂时间、泪液分泌量、眼表健康评分等)

支持向量机(SVM)

结合网格搜索优化

Kim S[30]

诊断泪道阻塞

泪囊造影图像数据集

 

CNN(VGG16)

Han F[31]

泪囊炎可解释性预测

眼部指标数据集

(泪膜客观散射指数、泪河高度、眨眼频率等)

 

深度堆叠网络(DSN)与SHAP 结合梯度增强分类器(GBC)

 

Wang Z[36]

监测泪液生物标志物

生物标志物数据集

(lgA、Lysozyme、MMP-9等)

随机森林(Random Forest)

Shimizu E[45]

估计泪膜破裂时间

眼表指标数据集

(泪膜破裂时间、眼表染色评分、泪液分泌量等)

CNN(Inception-v3)

Nowak R[50]

VR/MR辅助泪道阻塞手术治疗

临床病例资料数据集

 

VR/MR

 

 

4 AI在泪器疾病治疗中的作用

4.1 个性化治疗

       个性化治疗的关键在于AI能够整合不同来源的数据,深入挖掘其中的潜在信息,综合分析患者的病史、影像数据及临床表现,为患者推荐个性化治疗方案并预测患者对不同治疗方案的响应效果[48]。例如,AI技术可以根据患者的具体数据,精准选择最适合的药物或手术方式以优化治疗效果。此外,在动态调整方面,通过持续监测患者的泪道状态和症状变化,AI能够提供实时的数据分析,辅助医生调整治疗方案,不仅提升了治疗的精确度,还能够及时响应患者的治疗反馈,进一步改善治疗结果。个性化药物剂量的调整也是AI技术的重要应用之一。AI能够根据患者的个体差异以及治疗反应数据,推荐最适合的药物剂量,从而最大限度地提高疗效,并使可能的不良反应最小化[49]。通过精细的个性化管理,AI能够为患者提供更加精准和高效的医疗服务。这种创新的治疗方法不仅提高了治疗的安全性,还为个性化精准医疗提供了有力支持。

4.2 手术决策支持

       泪器系统解剖结构复杂且相互关联,需精确定位和高效处理,这与外科手术中的决策过程高度一致。因此,AI在外科领域的应用,尤其是在影像分析和手术决策支持方面,为泪器疾病的治疗提供了有力的理论基础。术前,AI通过深入分析患者的影像数据,如眼部CT或MRI扫描,为泪道和泪腺手术提供精确的路径和操作建议,从而优化手术效果并降低并发症风险。此外,AI通过虚拟现实(virtual reality,VR)和混合现实(mixed reality,MR)技术合成计算机视觉,进行术前模拟,预测泪道及泪腺手术中的潜在风险和最佳操作步骤,同时最大程度地模拟并优化手术效果,Nowak等[50]通过对一例创伤导致严重泪道梗阻的患者进行前瞻性研究,利用VR进行术前规划和MR辅助术中指导,显著改善了手术的精确性和效果。研究结果表明,VR和MR技术能够有效帮助外科医生理解复杂解剖结构,提高手术成功率,展示了其在泪道手术中的潜在应用价值。另外,AI利用ML算法处理患者的病史、影像数据和实验室结果,能够有效预测手术风险,并识别出潜在的高风险患者,从而辅助医生制定个性化的手术计划[51]。例如,对于泪道阻塞或其他复杂泪道疾病,AI能够推荐最合适的手术方案,如人工泪管置入或内镜手术。术中,AI通过实时分析手术视频和生理数据,及时识别并预警可能的并发症,如出血或泪道损伤,以便手术过程中及时调整。术后,AI通过分析术后恢复数据,预测可能的并发症和恢复情况,Qu等[52]利用多通道CNN的眼部模型预测眼部手术效果,进而制定提高眼部手术效果的方案,降低术后并发症的发生率,进而优化后续治疗方案,提高术后护理效果[53]

4.3 肿瘤放射组学

       泪腺肿瘤的放射治疗要求精确的肿瘤定位和剂量分布优化,以确保治疗效果的最大化和不良反应的最小化。AI通过分析CT、MRI等影像资料,能够自动识别和分割肿瘤组织[54]。这种智能化的分割能力不仅提高了肿瘤定位的准确性,还大大减少了人工操作带来的误差。此外,AI还可以从影像中提取关键的生物标志物,为肿瘤的性质和发展阶段提供更多的信息,从而帮助医生进行更精确的诊断和治疗计划制定。利用ML算法,AI能够优化放射治疗计划[55],通过模拟不同的剂量分布和照射角度,AI能够确保肿瘤组织接受足够的辐射剂量,同时有效保护周围正常组织,减少不良反应。此外,AI技术还能够根据肿瘤的影像特征和临床数据预测治疗效果和患者预后。通过对历史病例的分析,AI可以识别出不同患者对放疗的反应模式,从而为医生提供个性化的治疗建议。例如,针对特定类型的泪腺肿瘤,AI可以推荐最佳的剂量方案和治疗周期,以提高疗效并降低复发风险。
AI介导的自适应放疗(adaptive radiotherapy)更是展示了其创新性[7]。通过实时监测患者的肿瘤变化和正常组织的反应,AI能够及时调整放射治疗计划,以应对肿瘤的生长、形状变化或位置移动。这种动态调整能力不仅提升了放疗的精准度,还使得患者在整个治疗过程中的辐射暴露降到最低,极大地提升了治疗效果和患者的生活质量。

5 AI在泪器疾病诊疗中的挑战

5.1 多模态数据融合的复杂性

       泪器疾病的有效诊疗通常依赖于多种类型的数据,这些数据包括影像学检查(如泪道造影、CT和MRI)、生物标志物(如泪液成分分析和血液检测),以及临床症状和患者历史记录。这些数据在格式、维度和特征上存在显著差异,而将这些异质数据有效地整合到一个统一的AI模型中,是当前研究中的一大挑战。AI需要克服数据的不一致性、缺失和不匹配等问题,同时确保融合后的信息能够全面而准确地反映疾病的多方面特征,从而支持全面的疾病评估和制定个性化的治疗方案[56]。这一过程的复杂性对技术要求极高,同时也为技术的创新提供了机遇,推动了相关领域的进一步发展。有效的多模态数据融合将大幅提升AI在泪器疾病诊疗中的应用效果,有助于更全面、准确地评估疾病并提供治疗方案,为未来的研究和临床实践奠定基础。

5.2 模型泛化能力

       AI模型的泛化能力,即其在不同人群和环境中的表现,体现了AI模型在处理新见数据时保持高性能的能力。如何确保AI模型在各种临床情况下的有效性和可靠性是当前研究的重点。临床环境中往往会遇到多样化的病例,而模型的主要任务是从训练数据中学习到高效准确的模式,并在实际应用中对不同患者表现出一致的准确性。具体来说,AI在泪器疾病的诊断和预测任务中,如果仅在训练数据上表现优异,而在实际临床数据中表现不佳,则说明模型的泛化能力不足。这可能是训练数据的样本不足、数据多样性不足或数据标注不准确等因素所致[56]。为提高模型的泛化能力,需要大量多样化的训练数据和有效的正则化技术,以减少模型对训练数据的过度拟合。此外,为增强模型的泛化能力,AI研究人员需要探索跨区域、跨人群的数据集,确保模型能够适应不同的泪器疾病表现和患者特征。

5.3 实时预测和动态调整

       在泪器疾病的治疗过程中,AI需要具备实时预测患者反应的能力,并根据实时数据动态调整治疗策略。这要求AI不仅具备高效的数据处理能力,还需要具备实时学习和更新的能力。如何在临床应用中实时监测病情变化,并根据新数据动态调整诊疗策略是AI面临的一大挑战。首先,AI系统需要处理不同来源的数据,如患者的临床症状、影像数据和生物标志物,这要求其能够迅速更新和分析数据,以提供及时的诊断和治疗建议[57]。其次,AI的预测模型必须具备高度的适应性,以基于实时数据进行动态预测[7]。这意味着模型需要不断训练和更新,以反映疾病的最新状态,并做出准确的预测。此外,AI系统能够基于预测结果动态调整治疗方案,包括对药物剂量、治疗频率或干预措施的调整。在动态调整过程中,AI还需整合临床医生的反馈。因此,AI系统必须设计有效的反馈机制,以便依据实时数据和预测结果进行相应的干预和调整。克服这些挑战不仅能提高AI在泪器疾病诊疗中的应用效果,还能使治疗过程更加精准和个性化,以适应不同患者病情的变化。 

6 未来发展方向

6.1 新兴技术与融合

       6.1.1 AI与大数据、物联网结合实现泪器疾病远程监护和医疗
       AI与大数据的结合为泪器疾病的诊疗提供了前所未有的机遇和挑战。大数据技术能够高效收集和分析大量的患者数据,包括临床记录、影像数据、基因信息等,形成全面的疾病特征数据库[58]。这些数据不仅展示了泪器疾病的多样性,还揭示了潜在的病理模式,极大地丰富了疾病的理解和诊断依据。在此基础上,AI算法特别是ML和DL技术,在处理和分析这些庞大的数据集时发挥了关键作用[59]。研究表明,AI结合大数据、物联网(internet of things,IoT)分析能显著提升泪器疾病的诊断准确性和治疗效果。
       IoT技术通过将不同类型的传感器和医疗设备,如生物传感器、可穿戴设备、远程监测设备等连接到网络,实现了对患者健康状况的实时监测[60-61]。在泪器疾病的管理中,IoT设备能够实时监测、记录和传输多个关键的眼部生理指标,包括泪液分泌量、瞬目频率、眼表干燥程度等。通过对这些实时数据进行智能分析,AI能够快速识别出病情的变化趋势,并生成个性化的管理建议。此外,AI与IoT及大数据的结合支持远程监控和干预,提高了疾病的管理效率,为患者提供了更加便捷高效的医疗服务。
       6.1.2 AI与虚拟现实/增强现实结合辅助泪器疾病术前规划
       VR和AR技术在医疗领域中展现出了巨大的应用潜力[54,62]。将AI与VR/AR技术结合,有望在泪器疾病的诊疗中实现前所未有的创新。AI驱动的AR技术通过实时增强视图,为医生在手术过程中提供了更加精准的辅助[63]。此外,AR技术可以将虚拟的解剖结构图层叠加在实际的视野中,使医生能够清晰地识别和定位泪器各部分解剖结构[50]。增强现实视图能够帮助医生准确了解病变区域,提高手术的精确度,减少人为误差,从而改善手术效果[64]。例如,AI可以分析患者的泪道影像数据,生成详细的三维模型,并将三维模型在手术过程中以AR形式投影到医生的视野中。这类技术的应用不仅提高了对复杂泪器结构的认识,也降低了手术中的不可预测性和风险。 
       此外,VR技术在手术前模拟训练和规划中也发挥了重要作用[65]。未来,通过AI与VR的结合模拟泪器疾病的病理生理状态,能帮助医生在进入实际手术前进行详细的虚拟训练,提高手术的成功率[1]。AI算法创建的高精度虚拟模型能够真实地再现患者的病情,使医生可以在虚拟环境中进行手术步骤的预演。这种训练方式不仅有助于医生全面了解疾病特征,还能够提高手术技能和流畅度,优化手术方案。图2展示了AI加持下的VR辅助医疗的工作流程,这种技术的融合和进步将对疾病的诊疗产生深远的影响,推动医疗技术尤其是眼科精细操作的进一步发展。
图 2 AI与VR结合辅助术前规划的工作流程[62]
Figure 2 Workflow of AI and VR integration for assisted preoperative planning
(I)通过对抗网络(generative adversarial networks, GANs)和扩散模型相结合,促进视觉可视化增强。(II)AI与VR结合辅助医疗数据的处理。 (III)VR与AI结合指导相位识别,辅助术前规划,通过视觉问题回答(uncertainty quantification in vision, UQA)和视觉问题本地化回答(uncertainty quantification and localization algorithm, UQLA),促进互动合作,提高患者的理解和依从性。
(I) Enhancing visual visualization through the integration of Generative Adversarial Networks (GANs) and Diffusion Models. (II) AI and VR integration for assisting in medical data processing. (III) VR and AI integration for guiding phase recognition and supporting preoperative planning through uncertainty quantification in vision (UQA) and uncertainty quantification and localization algorithm (UQLA), promoting interactive collaboration to improve patient understanding and compliance.
       6.1.3 AI与生物传感器结合推动泪器疾病的个性化治疗
       生物传感器技术在实时监测生理参数方面展现出了显著的应用潜力,结合AI后,其有望提供更为精确和全面的疾病监测方案[61]。生物传感器能够对泪液成分、泪道压力等关键指标进行实时检测和分析,而AI的集成则进一步增强了这些数据的应用价值。随着AI技术的发展,AI驱动的生物传感器使实时分析泪液中多种生化指标如泪液黏度、蛋白质浓度和炎症标志物等成为可能。Lu等[66]研发了一种无线便携式泪液分析传感器,灵敏检测泪液中与干眼综合征(dry eye syndrome,DES)相关的四种细胞因子干扰素-γ(interferon-gamma,IFN-γ)、白细胞介素-6(interleukin-6,IL-6)、肿瘤坏死因子-α(tumor necrosis factor-alpha,TNF-α)、基质金属蛋白酶-9(matrix metalloproteinase-9,MMP-9),该纳米传感器的检测策略基于DNA四面体框架(DNA tetrahedron framework,DTF)与三维杂交链反应(3D hybridization chain reaction,3D-HCR)放大器的偶联(图3)。在这项新研究中,该团队利用DTF有效地捕获了具有可控多分支臂的3D-HCR产物。结果表明,与单一生物标志物诊断相比,基于多种生物标志物的传感器的诊断准确性提高了约16%。
图 3 检测泪液中细胞因子(IFN-γ、IL-6、TNF-α、MMP-9)的新型生物传感器技术原理[66]
Figure 3 The principle of a novel biosensor technology for detecting cytokines (IFN-γ, IL-6, TNF-α, MMP-9) in tear fluid
(A) 利用无线便携式电化学传感器对泪液中细胞因子(如IFN-𝛾、IL-6、TNF-𝛼、MMP-9) 进行分析,实现眼部疾病的诊断。(B)所构建的无线便携式泪液分析传感器原理图设计。该传感器由电极芯片和便携式电化学工作站组成。电极芯片的分层设计从上到下依次为绝缘层、反应层、导体层、衬底层。该工作站有5个主要单元:蓝牙芯片、数据处理芯片、电源管理芯片、MS02生物芯片和充电芯片。(C)反应层电极界面上的DNA四面体框架(DNA tetrahedron frameworks,DTFs)示意图,用于组装多维HCR产物,进而用于细胞因子的敏感测定。(D)便携式电化学工作站的模拟图。(E)毛细管泪液收集器。(F)通过毛细管泪液收集器将样品装载到电极芯片上。(G)用于细胞因子监测的无线便携式泪液分析传感器。
(A) Utilization of a wireless portable electrochemical sensor for the analysis of cytokines (such as IFN-γ, IL-6, TNF-α, MMP-9) in tear fluid for the diagnosis of ocular diseases. (B) Schematic design of the wireless portable tear fluid analysis sensor. The sensor consists of an electrode chip and a portable electrochemical workstation. The layered design of the electrode chip includes an insulating layer, a reaction layer, a conductor layer, and a substrate layer, from top to bottom. The workstation comprises five main units: a Bluetooth chip, a data processing chip, a power management chip, an MS02 biosensor chip, and a charging chip. (C) Schematic diagram of the DNA tetrahedron frameworks (DTFs) on the reaction layer electrode interface, used for assembling multidimensional HCR products for sensitive cytokine detection. (D) Simulated diagram of the portable electrochemical workstation.(E) Capillary tear fluid collector. (F)which loads samples onto the electrode chip using capillary action. (G) Wireless portable tear fluid analysis sensor for cytokine monitoring.

       此外,通过高级的数据分析和模式识别算法,AI能够处理复杂的传感器数据,并实时生成关于泪器疾病的动态信息。这种信息不仅有助于疾病状态的持续监测,还能在早期检测出病情的细微变化,从而实现更早的干预和治疗[67–69]。例如,利用AI对历史数据深入分析和模式识别的能力,生物传感器能够预测病情的潜在波动,提供预警信息,辅助医生及时调整治疗方案,优化治疗效果[70]
6.1.4 AI与纳米技术结合促进泪器疾病诊疗的发展
       近年来,纳米技术在医学领域的应用引起了广泛关注,尤其是其与AI的结合[71],有望显著推进泪器疾病诊疗的进步。纳米技术在药物递送系统中的应用显著提升了药物的靶向性和治疗效果。通过ML算法,AI有望优化纳米药物递送系统的设计,辅助纳米药物的筛选和开发[72]。此外,AI能够精确预测纳米颗粒在泪道和泪腺内的行为,从而设计出适合特定病变区域的递送系统。使药物能够精准靶向病变部位,提高局部药物浓度并减少对健康组织的影响,实现更高效的药物输送。
       此外,纳米传感器在泪液成分分析及泪器疾病的实时监测中发挥着重要作用。凭借其高灵敏度和高特异性,与AI结合的纳米传感器有望检测出泪液中的微量生物标志物和病理变化,并通过深度分析数据,提取有价值的信息,提供早期预警。基于硅纳米线场效应的生物传感器能够捕获特异性的靶分子,从而感受由靶分子电荷引起的导电性能变化,具有灵敏度高、响应迅速、所需样本量少、无需标记等特点[73-74]。Lu等[75]创新性地采用光学校准的方法克服了由于芯片制造和功能修饰过程中引入的传感器间差异,运用低盐溶液置换法解决了生物传感器在实际临床环境中因体液样品成分复杂而造成的应用障碍,首创研发了一款用于定量检测泪液MMP-9蛋白浓度的硅纳米线生物传感器芯片,原理如图4所示。该生物传感器与传统的酶联免疫吸附测定法之间具有高度一致性,诊断灵敏度为 86.96%,特异度为 90%。纳米技术的精准药物递送和高效检测能力与AI的智能分析和预测功能相结合,有望为泪器疾病的早期诊断和精准治疗提供新的可能性。
图 4 检测基质金属蛋白酶9 (MMP-9)的硅纳米线场效应晶体管(SiNW FET)生物传感器的测试程序[75]
Figure 4 Testing procedure of the silicon nanowire field-effect transistor (SiNW FET) biosensor for the detection of matrix metalloproteinase 9 (MMP-9)
(A)用于即时检测的便携式SiNW检测系统。(B)生物传感器功能化方案:(I)可控质量SiNW器件生产;(II)器件的表面修饰;(III)使用光学方法校准设备。(C)生物传感器的测试程序:(I)用特制的泪液采集装置采集样本;(II)样本与100 μL的100 nmol/L的磷酸盐缓冲盐水(Phosphate Buffered Saline, PBS)混合;(III)离心制取上清液,并(IV)滴在SiNW装置上(V)连续记录响应。
(A) Portable SiNW detection system for real-time monitoring. (B) Biosensor functionalization scheme: (I) Controlled fabrication of SiNW devices; (II) Surface modification of the devices; (III) Calibration of the equipment using optical methods. (C) Biosensor testing procedure: (I) Collection of samples using a custom-designed tear collection device; (II) Mixing the samples with 100 μL of 100 nmol/L Phosphate Buffered Saline (PBS); (III) Centrifugation to obtain the supernatant, and (IV) Dropping the supernatant onto the SiNW device; (V) Continuous recording of the response.

6.2 多学科合作提高泪器疾病诊疗的有效性

       AI的广泛应用需要医学专家、计算机科学家和数据科学家的多学科合作,以推动技术的发展。整合临床经验与数据科学是关键步骤[76–78]。医学专家提供丰富的临床知识和病理数据以训练和优化AI算法,数据科学家确保数据的高质量和准确标注,进一步提升算法的诊断准确性和实用性。这种紧密的合作能够使AI更好地适应实际的临床需求,从而提高其在疾病诊疗中的有效性。在跨领域技术创新方面,工程师和技术开发者致力于将AI技术整合进医疗设备中,涉及硬件设备的设计以及软件平台的开发等。通过多学科合作,AI技术有望在实际医疗环境中得到更广泛的应用,并提高AI的使用效率和便捷性。多学科合作确保了AI不仅在理论上有效,而且在实际临床环境中表现优越[79–81]。此外,通过跨学科团队的努力,AI算法可以不断优化和调整,以适应新的医疗发现和临床反馈,从而保持技术的前沿性和实用性,最终实现更高效、更精准的泪器疾病诊疗。

7 结语

       本综述深入探讨了AI在泪器疾病诊疗中的应用,全面分析了其存在的挑战与机遇。AI技术在泪器疾病的早期筛查、精准诊疗和预后预测方面展现出了显著的优势。然而,其仍需克服多模态数据融合的复杂性、模型泛化能力的局限以及实时预测和动态调整的需求等诸多困难。针对这些问题,本文旨在为未来研究者提供创新性的观点,促进AI在泪器疾病中的有效应用,从而推动泪器疾病诊疗的创新性发展。

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。
1、Agrahari V, Choonara YE, Mosharraf M, et al. The role of artificial intelligence and machine learning in accelerating the discovery and development of nanomedicine[J]. Pharm Res, 2024, 41(12): 2289-2297. DOI:10.1007/s11095-024-03798-9. Agrahari V, Choonara YE, Mosharraf M, et al. The role of artificial intelligence and machine learning in accelerating the discovery and development of nanomedicine[J]. Pharm Res, 2024, 41(12): 2289-2297. DOI:10.1007/s11095-024-03798-9.
2、Othman D, Kaleem A. The intraoperative role of artificial intelligence within general surgery: a systematic review[J]. Cureus, 2024, 16(11): e73006. DOI:10.7759/cureus.73006. Othman D, Kaleem A. The intraoperative role of artificial intelligence within general surgery: a systematic review[J]. Cureus, 2024, 16(11): e73006. DOI:10.7759/cureus.73006.
3、Deo RC. Machine learning in medicine[J]. Circulation, 2015, 132(20): 1920-1930. DOI:10.1161/CIRCULATIONAHA.115.001593.Deo RC. Machine learning in medicine[J]. Circulation, 2015, 132(20): 1920-1930. DOI:10.1161/CIRCULATIONAHA.115.001593.
4、Jiang Y, Yang M, Wang S, et al. Emerging role of deep learning-based artificial intelligence in tumor pathology[J]. Cancer Commun, 2020, 40(4): 154-166. DOI:10.1002/cac2.12012. Jiang Y, Yang M, Wang S, et al. Emerging role of deep learning-based artificial intelligence in tumor pathology[J]. Cancer Commun, 2020, 40(4): 154-166. DOI:10.1002/cac2.12012.
5、Leandro I, Lorenzo B, Aleksandar M, et al. OCT-based deep-learning models for the identification of retinal key signs[J]. Sci Rep, 2023, 13(1): 14628. DOI:10.1038/s41598-023-41362-4.Leandro I, Lorenzo B, Aleksandar M, et al. OCT-based deep-learning models for the identification of retinal key signs[J]. Sci Rep, 2023, 13(1): 14628. DOI:10.1038/s41598-023-41362-4.
6、Gulshan V, Peng L, Coram M, et al. Development and validation of a deep learning algorithm for detection of diabetic retinopathy in retinal fundus photographs[J]. JAMA, 2016, 316(22): 2402-2410. DOI:10.1001/jama.2016.17216.Gulshan V, Peng L, Coram M, et al. Development and validation of a deep learning algorithm for detection of diabetic retinopathy in retinal fundus photographs[J]. JAMA, 2016, 316(22): 2402-2410. DOI:10.1001/jama.2016.17216.
7、Ting DSW, Peng L, Varadarajan AV, et al. Deep learning in ophthalmology: The technical and clinical considerations[J]. Prog Retin Eye Res, 2019, 72: 100759. DOI:10.1016/j.preteyeres.2019.04.003. Ting DSW, Peng L, Varadarajan AV, et al. Deep learning in ophthalmology: The technical and clinical considerations[J]. Prog Retin Eye Res, 2019, 72: 100759. DOI:10.1016/j.preteyeres.2019.04.003.
8、Zhang YY, Zhao H, Lin JY, et al. Artificial intelligence to detect meibomian gland dysfunction from in-vivo laser confocal microscopy[J]. Front Med, 2021, 8: 774344. DOI:10.3389/fmed.2021.774344. Zhang YY, Zhao H, Lin JY, et al. Artificial intelligence to detect meibomian gland dysfunction from in-vivo laser confocal microscopy[J]. Front Med, 2021, 8: 774344. DOI:10.3389/fmed.2021.774344.
9、Deng J, Qin Y. Current status, hotspots, and prospects of artificial intelligence in ophthalmology: a bibliometric analysis (2003-2023)[J]. Ophthalmic Epidemiol, 2024: 1-14. DOI:10.1080/09286586.2024.2373956. Deng J, Qin Y. Current status, hotspots, and prospects of artificial intelligence in ophthalmology: a bibliometric analysis (2003-2023)[J]. Ophthalmic Epidemiol, 2024: 1-14. DOI:10.1080/09286586.2024.2373956.
10、Quaranta-Leoni FM, Fiorino MG, Serricchio F, et al. Management of proximal lacrimal obstructions: a rationale[J]. Acta Ophthalmol, 2021, 99(4): e569-e575. DOI:10.1111/aos.14632. Quaranta-Leoni FM, Fiorino MG, Serricchio F, et al. Management of proximal lacrimal obstructions: a rationale[J]. Acta Ophthalmol, 2021, 99(4): e569-e575. DOI:10.1111/aos.14632.
11、Kim JS, Al-Lozi A, Leyngold IM. Malignant orbital tumors: current approach to diagnosis and management[J]. Curr Ophthalmol Rep, 2021, 9(1): 16-24. DOI:10.1007/s40135-020-00262-w.Kim JS, Al-Lozi A, Leyngold IM. Malignant orbital tumors: current approach to diagnosis and management[J]. Curr Ophthalmol Rep, 2021, 9(1): 16-24. DOI:10.1007/s40135-020-00262-w.
12、Kim YH, Graham AD, Li W, et al. Human lacrimal production rate and wetted length of modified schirmer’s tear test strips[J]. Transl Vis Sci Technol, 2019, 8(3): 40. DOI:10.1167/tvst.8.3.40. Kim YH, Graham AD, Li W, et al. Human lacrimal production rate and wetted length of modified schirmer’s tear test strips[J]. Transl Vis Sci Technol, 2019, 8(3): 40. DOI:10.1167/tvst.8.3.40.
13、Wulu%20JA%2C%20Spiegel%20JH.%20Is%20a%20schirmer%E2%80%99s%20test%20necessary%20before%20blepharoplasty%3F%5BJ%5D.%20Laryngoscope%2C%202019%2C%20129(5)%3A%201021-1022.%20DOI%3A10.1002%2Flary.27446.Wulu%20JA%2C%20Spiegel%20JH.%20Is%20a%20schirmer%E2%80%99s%20test%20necessary%20before%20blepharoplasty%3F%5BJ%5D.%20Laryngoscope%2C%202019%2C%20129(5)%3A%201021-1022.%20DOI%3A10.1002%2Flary.27446.
14、Nakamura J, Kamao T, Mitani A, et al. Accuracy of the lacrimal syringing test in relation to dacryocystography and dacryoendoscopy[J]. Clin Ophthalmol, 2023, 17: 1277-1285. DOI:10.2147/OPTH.S409662.Nakamura J, Kamao T, Mitani A, et al. Accuracy of the lacrimal syringing test in relation to dacryocystography and dacryoendoscopy[J]. Clin Ophthalmol, 2023, 17: 1277-1285. DOI:10.2147/OPTH.S409662.
15、Maliborski%20A%2C%20R%C3%B3%C5%BCycki%20R.%20Diagnostic%20imaging%20of%20the%20nasolacrimal%20drainage%20system.%20Part%20I.%20Radiological%20anatomy%20of%20lacrimal%20pathways.%20Physiology%20of%20tear%20secretion%20and%20tear%20outflow%5BJ%5D.%20Med%20Sci%20Monit%2C%202014%2C%2020%3A%20628-638.%20DOI%3A10.12659%2FMSM.890098.%20Maliborski%20A%2C%20R%C3%B3%C5%BCycki%20R.%20Diagnostic%20imaging%20of%20the%20nasolacrimal%20drainage%20system.%20Part%20I.%20Radiological%20anatomy%20of%20lacrimal%20pathways.%20Physiology%20of%20tear%20secretion%20and%20tear%20outflow%5BJ%5D.%20Med%20Sci%20Monit%2C%202014%2C%2020%3A%20628-638.%20DOI%3A10.12659%2FMSM.890098.%20
16、Park DH, Connor KM, Lambris JD. The challenges and promise of complement therapeutics for ocular diseases[J]. Front Immunol, 2019, 10: 1007. DOI:10.3389/fimmu.2019.01007. Park DH, Connor KM, Lambris JD. The challenges and promise of complement therapeutics for ocular diseases[J]. Front Immunol, 2019, 10: 1007. DOI:10.3389/fimmu.2019.01007.
17、Takayanagi H, Hayashi R. Status and prospects for the development of regenerative therapies for corneal and ocular diseases[J]. Regen Ther, 2024, 26: 819-825. DOI:10.1016/j.reth.2024.09.001.Takayanagi H, Hayashi R. Status and prospects for the development of regenerative therapies for corneal and ocular diseases[J]. Regen Ther, 2024, 26: 819-825. DOI:10.1016/j.reth.2024.09.001.
18、Su TY, Ho WT, Lu CY, et al. Correlations among ocular surface temperature difference value, the tear meniscus height, Schirmer’s test and fluorescein tear film break up time[J]. Br J Ophthalmol, 2015, 99(4): 482-487. DOI:10.1136/bjophthalmol-2014-305183.Su TY, Ho WT, Lu CY, et al. Correlations among ocular surface temperature difference value, the tear meniscus height, Schirmer’s test and fluorescein tear film break up time[J]. Br J Ophthalmol, 2015, 99(4): 482-487. DOI:10.1136/bjophthalmol-2014-305183.
19、Imamura H, Tabuchi H, Nagasato D, et al. Automatic screening of tear meniscus from lacrimal duct obstructions using anterior segment optical coherence tomography images by deep learning[J]. Graefes Arch Clin Exp Ophthalmol, 2021, 259(6): 1569-1577. DOI:10.1007/s00417-021-05078-3. Imamura H, Tabuchi H, Nagasato D, et al. Automatic screening of tear meniscus from lacrimal duct obstructions using anterior segment optical coherence tomography images by deep learning[J]. Graefes Arch Clin Exp Ophthalmol, 2021, 259(6): 1569-1577. DOI:10.1007/s00417-021-05078-3.
20、Song X, Li L, Han F, et al. Noninvasive machine learning screening model for dacryocystitis based on ocular surface indicators[J]. J Craniofac Surg, 2022, 33(1): e23-e28. DOI:10.1097/SCS.0000000000007863.Song X, Li L, Han F, et al. Noninvasive machine learning screening model for dacryocystitis based on ocular surface indicators[J]. J Craniofac Surg, 2022, 33(1): e23-e28. DOI:10.1097/SCS.0000000000007863.
21、 Benou A, Veksler R, Friedman A, et al. Ensemble of expert deep neural networks for spatio-temporal denoising of contrast-enhanced MRI sequences[J]. Med Image Anal, 2017, 42: 145-159. DOI:10.1016/j.media.2017.07.006. Benou A, Veksler R, Friedman A, et al. Ensemble of expert deep neural networks for spatio-temporal denoising of contrast-enhanced MRI sequences[J]. Med Image Anal, 2017, 42: 145-159. DOI:10.1016/j.media.2017.07.006.
22、Plassard AJ, Davis LT, Newton AT, et al. Learning implicit brain MRI manifolds with deep learning[C]//Medical Imaging 2018: Image Processing. February 10-15, 2018. Houston, USA. SPIE, 2018: 10.1117/12.2293515. DOI:10.1117/12.2293515. Plassard AJ, Davis LT, Newton AT, et al. Learning implicit brain MRI manifolds with deep learning[C]//Medical Imaging 2018: Image Processing. February 10-15, 2018. Houston, USA. SPIE, 2018: 10.1117/12.2293515. DOI:10.1117/12.2293515.
23、Chaudhari AS, Fang Z, Kogan F, et al. Super-resolution musculoskeletal MRI using deep learning[J]. Magn Reson Med, 2018, 80(5): 2139-2154. DOI:10.1002/mrm.27178. Chaudhari AS, Fang Z, Kogan F, et al. Super-resolution musculoskeletal MRI using deep learning[J]. Magn Reson Med, 2018, 80(5): 2139-2154. DOI:10.1002/mrm.27178.
24、Liu C, Wu X, Yu X, et al. Fusing multi-scale information in convolution network for MR image super-resolution reconstruction[J]. Biomed Eng Online, 2018, 17(1): 114. DOI:10.1186/s12938-018-0546-9.Liu C, Wu X, Yu X, et al. Fusing multi-scale information in convolution network for MR image super-resolution reconstruction[J]. Biomed Eng Online, 2018, 17(1): 114. DOI:10.1186/s12938-018-0546-9.
25、Hou R, Zhou D, Nie R, et al. Brain CT and MRI medical image fusion using convolutional neural networks and a dual-channel spiking cortical model[J]. Med Biol Eng Comput, 2019, 57(4): 887-900. DOI:10.1007/s11517-018-1935-8. Hou R, Zhou D, Nie R, et al. Brain CT and MRI medical image fusion using convolutional neural networks and a dual-channel spiking cortical model[J]. Med Biol Eng Comput, 2019, 57(4): 887-900. DOI:10.1007/s11517-018-1935-8.
26、Noble JA. Ultrasound image segmentation and tissue characterization[J]. Proc Inst Mech Eng H, 2010, 224(2): 307-316. DOI:10.1243/09544119JEIM604. Noble JA. Ultrasound image segmentation and tissue characterization[J]. Proc Inst Mech Eng H, 2010, 224(2): 307-316. DOI:10.1243/09544119JEIM604.
27、Hamwood J, Schmutz B, Collins MJ, et al. A deep learning method for automatic segmentation of the bony orbit in MRI and CT images[J]. Sci Rep, 2021, 11(1): 13693. DOI:10.1038/s41598-021-93227-3.Hamwood J, Schmutz B, Collins MJ, et al. A deep learning method for automatic segmentation of the bony orbit in MRI and CT images[J]. Sci Rep, 2021, 11(1): 13693. DOI:10.1038/s41598-021-93227-3.
28、Fu R, Leader JK, Pradeep T, et al. Automated delineation of orbital abscess depicted on CT scan using deep learning[J]. Med Phys, 2021, 48(7): 3721-3729. DOI:10.1002/mp.14907. Fu R, Leader JK, Pradeep T, et al. Automated delineation of orbital abscess depicted on CT scan using deep learning[J]. Med Phys, 2021, 48(7): 3721-3729. DOI:10.1002/mp.14907.
29、Brown JM, Peter Campbell J, Beers A, et al. Automated diagnosis of plus disease in retinopathy of prematurity using deep convolutional neural networks[J]. JAMA Ophthalmol, 2018, 136(7): 803-810. DOI:10.1001/jamaophthalmol.2018.1934. Brown JM, Peter Campbell J, Beers A, et al. Automated diagnosis of plus disease in retinopathy of prematurity using deep convolutional neural networks[J]. JAMA Ophthalmol, 2018, 136(7): 803-810. DOI:10.1001/jamaophthalmol.2018.1934.
30、Kim S, Lee H, Roh HG, et al. Using artificial intelligence to diagnose lacrimal passage obstructions based on dacryocystography images[J]. J Craniofac Surg, 2024. DOI:10.1097/SCS.0000000000010829. Kim S, Lee H, Roh HG, et al. Using artificial intelligence to diagnose lacrimal passage obstructions based on dacryocystography images[J]. J Craniofac Surg, 2024. DOI:10.1097/SCS.0000000000010829.
31、Han F, Liao S, Song X, et al. Explainable prediction of dacryocystitis from noninvasive ocular indicators using deep stacked network and the shapley additive explanations approach[J]. J Craniofac Surg, 2022, 33(4): e350-e355. DOI:10.1097/SCS.0000000000008059. Han F, Liao S, Song X, et al. Explainable prediction of dacryocystitis from noninvasive ocular indicators using deep stacked network and the shapley additive explanations approach[J]. J Craniofac Surg, 2022, 33(4): e350-e355. DOI:10.1097/SCS.0000000000008059.
32、Ali MJ, Singh S. Optical coherence tomography and the proximal lacrimal drainage system: a major review[J]. Graefes Arch Clin Exp Ophthalmol, 2021, 259(11): 3197-3208. DOI:10.1007/s00417-021-05175-3. Ali MJ, Singh S. Optical coherence tomography and the proximal lacrimal drainage system: a major review[J]. Graefes Arch Clin Exp Ophthalmol, 2021, 259(11): 3197-3208. DOI:10.1007/s00417-021-05175-3.
33、Cai Y, Zhang X, Cao J, et al. Application of artificial intelligence in oculoplastics[J]. Clin Dermatol, 2024, 42(3): 259-267. DOI:10.1016/j.clindermatol.2023.12.019. Cai Y, Zhang X, Cao J, et al. Application of artificial intelligence in oculoplastics[J]. Clin Dermatol, 2024, 42(3): 259-267. DOI:10.1016/j.clindermatol.2023.12.019.
34、Park DI, Lew H, Lee SY. Tear meniscus measurement in nasolacrimal duct obstruction patients with Fourier-domain optical coherence tomography: novel three-point capture method[J]. Acta Ophthalmol, 2012, 90(8): 783-787. DOI:10.1111/j.1755-3768.2011.02183.x. Park DI, Lew H, Lee SY. Tear meniscus measurement in nasolacrimal duct obstruction patients with Fourier-domain optical coherence tomography: novel three-point capture method[J]. Acta Ophthalmol, 2012, 90(8): 783-787. DOI:10.1111/j.1755-3768.2011.02183.x.
35、Wang X, Fan X, Wu Y, et al. Rear 4-Min Schirmer test, a modified indicator of Schirmer test in diagnosing dry eye[J]. Sci Rep, 2022, 12(1): 6272. DOI:10.1038/s41598-022-09791-9. Wang X, Fan X, Wu Y, et al. Rear 4-Min Schirmer test, a modified indicator of Schirmer test in diagnosing dry eye[J]. Sci Rep, 2022, 12(1): 6272. DOI:10.1038/s41598-022-09791-9.
36、Wang Z, Dong Y, Sui X, et al. An artificial intelligence-assisted microfluidic colorimetric wearable sensor system for monitoring of key tear biomarkers[J]. NPJ Flex Electron, 2024, 8: 35. DOI:10.1038/s41528-024-00321-3. Wang Z, Dong Y, Sui X, et al. An artificial intelligence-assisted microfluidic colorimetric wearable sensor system for monitoring of key tear biomarkers[J]. NPJ Flex Electron, 2024, 8: 35. DOI:10.1038/s41528-024-00321-3.
37、Han SB, Yang HK, Hyon JY, et al. Association of dry eye disease with psychiatric or neurological disorders in elderly patients[J]. Clin Interv Aging, 2017, 12: 785-792. DOI:10.2147/CIA.S137580.Han SB, Yang HK, Hyon JY, et al. Association of dry eye disease with psychiatric or neurological disorders in elderly patients[J]. Clin Interv Aging, 2017, 12: 785-792. DOI:10.2147/CIA.S137580.
38、Stapleton F, Alves M, Bunya VY, et al. TFOS DEWS II epidemiology report[J]. Ocul Surf, 2017, 15(3): 334-365. DOI:10.1016/j.jtos.2017.05.003.Stapleton F, Alves M, Bunya VY, et al. TFOS DEWS II epidemiology report[J]. Ocul Surf, 2017, 15(3): 334-365. DOI:10.1016/j.jtos.2017.05.003.
39、Yamanishi R, Uchino M, Uchino Y, et al. Changes in distribution of dry eye diagnostic status among visual display terminal workers according to the revised criteria of the Asia dry eye society[J]. Cornea, 2020, 39(5): 578-583. DOI:10.1097/ICO.0000000000002218. Yamanishi R, Uchino M, Uchino Y, et al. Changes in distribution of dry eye diagnostic status among visual display terminal workers according to the revised criteria of the Asia dry eye society[J]. Cornea, 2020, 39(5): 578-583. DOI:10.1097/ICO.0000000000002218.
40、Nichols KK, Nichols JJ, Lynn Mitchell G. The lack of association between signs and symptoms in patients with dry eye disease[J]. Cornea, 2004, 23(8): 762-770. DOI:10.1097/01.ico.0000133997.07144.9e. Nichols KK, Nichols JJ, Lynn Mitchell G. The lack of association between signs and symptoms in patients with dry eye disease[J]. Cornea, 2004, 23(8): 762-770. DOI:10.1097/01.ico.0000133997.07144.9e.
41、Ong ES, Felix ER, Levitt RC, et al. Epidemiology of discordance between symptoms and signs of dry eye[J]. Br J Ophthalmol, 2018, 102(5): 674-679. DOI:10.1136/bjophthalmol-2017-310633. Ong ES, Felix ER, Levitt RC, et al. Epidemiology of discordance between symptoms and signs of dry eye[J]. Br J Ophthalmol, 2018, 102(5): 674-679. DOI:10.1136/bjophthalmol-2017-310633.
42、Vehof J, Sillevis Smitt-Kamminga N, Nibourg SA, et al. Predictors of discordance between symptoms and signs in dry eye disease[J]. Ophthalmology, 2017, 124(3): 280-286. DOI:10.1016/j.ophtha.2016.11.008. Vehof J, Sillevis Smitt-Kamminga N, Nibourg SA, et al. Predictors of discordance between symptoms and signs in dry eye disease[J]. Ophthalmology, 2017, 124(3): 280-286. DOI:10.1016/j.ophtha.2016.11.008.
43、Han SB, Hyon JY, Woo SJ, et al. Prevalence of dry eye disease in an elderly Korean population[J]. Arch Ophthalmol, 2011, 129(5): 633-638. DOI:10.1001/archophthalmol.2011.78.Han SB, Hyon JY, Woo SJ, et al. Prevalence of dry eye disease in an elderly Korean population[J]. Arch Ophthalmol, 2011, 129(5): 633-638. DOI:10.1001/archophthalmol.2011.78.
44、Nichols KK, Lynn Mitchell G, Zadnik K. The repeatability of clinical measurements of dry eye[J]. Cornea, 2004, 23(3): 272-285. DOI:10.1097/00003226-200404000-00010. Nichols KK, Lynn Mitchell G, Zadnik K. The repeatability of clinical measurements of dry eye[J]. Cornea, 2004, 23(3): 272-285. DOI:10.1097/00003226-200404000-00010.
45、Shimizu E, Ishikawa T, Tanji M, et al. Artificial intelligence to estimate the tear film breakup time and diagnose dry eye disease[J]. Sci Rep, 2023, 13: 5822. DOI:10.1038/s41598-023-33021-5.Shimizu E, Ishikawa T, Tanji M, et al. Artificial intelligence to estimate the tear film breakup time and diagnose dry eye disease[J]. Sci Rep, 2023, 13: 5822. DOI:10.1038/s41598-023-33021-5.
46、Tsubota K, Yokoi N, Shimazaki J, et al. New perspectives on dry eye definition and diagnosis: a consensus report by the Asia dry eye society[J]. Ocul Surf, 2017, 15(1): 65-76. DOI:10.1016/j.jtos.2016.09.003.Tsubota K, Yokoi N, Shimazaki J, et al. New perspectives on dry eye definition and diagnosis: a consensus report by the Asia dry eye society[J]. Ocul Surf, 2017, 15(1): 65-76. DOI:10.1016/j.jtos.2016.09.003.
47、Tsubota K, Yokoi N, Watanabe H, et al. A new perspective on dry eye classification: proposal by the Asia dry eye society[J]. Eye Contact Lens, 2020, 46 Suppl(1): S2-S13. DOI:10.1097/ICL.0000000000000643. Tsubota K, Yokoi N, Watanabe H, et al. A new perspective on dry eye classification: proposal by the Asia dry eye society[J]. Eye Contact Lens, 2020, 46 Suppl(1): S2-S13. DOI:10.1097/ICL.0000000000000643.
48、 Li YH, Li YL, Wei MY, et al. Innovation and challenges of artificial intelligence technology in personalized healthcare[J]. Sci Rep, 2024, 14(1): 18994. DOI:10.1038/s41598-024-70073-7. Li YH, Li YL, Wei MY, et al. Innovation and challenges of artificial intelligence technology in personalized healthcare[J]. Sci Rep, 2024, 14(1): 18994. DOI:10.1038/s41598-024-70073-7.
49、Chakravarty K, Antontsev V, Bundey Y, et al. Driving success in personalized medicine through AI-enabled computational modeling[J]. Drug Discov Today, 2021, 26(6): 1459-1465. DOI:10.1016/j.drudis.2021.02.007. Chakravarty K, Antontsev V, Bundey Y, et al. Driving success in personalized medicine through AI-enabled computational modeling[J]. Drug Discov Today, 2021, 26(6): 1459-1465. DOI:10.1016/j.drudis.2021.02.007.
50、Nowak%20R%2C%20Nowak-Gospodarowicz%20I%2C%20R%C4%99kas%20M%2C%20et%20al.%20Virtual%20reality%20and%20mixed%20reality-assisted%20endoscopic%20DCR%20in%20extremely%20complex%20lacrimal%20obstructions%5BJ%5D.%20Laryngoscope%2C%202024%2C%20134(8)%3A%203508-3515.%20DOI%3A10.1002%2Flary.31399.%20Nowak%20R%2C%20Nowak-Gospodarowicz%20I%2C%20R%C4%99kas%20M%2C%20et%20al.%20Virtual%20reality%20and%20mixed%20reality-assisted%20endoscopic%20DCR%20in%20extremely%20complex%20lacrimal%20obstructions%5BJ%5D.%20Laryngoscope%2C%202024%2C%20134(8)%3A%203508-3515.%20DOI%3A10.1002%2Flary.31399.%20
51、Loftus TJ, Tighe PJ, Filiberto AC, et al. Artificial intelligence and surgical decision-making[J]. JAMA Surg, 2020, 155(2): 148-158. DOI:10.1001/jamasurg.2019.4917.Loftus TJ, Tighe PJ, Filiberto AC, et al. Artificial intelligence and surgical decision-making[J]. JAMA Surg, 2020, 155(2): 148-158. DOI:10.1001/jamasurg.2019.4917.
52、YixinQu, BingyingLin, ShuilingLi, et al. Effect of multichannel convolutional neural network-based model on the repair and aesthetic effect of eye plastic surgery patients[J]. Comput Math Methods Med, 2022, 2022: 5315146. DOI:10.1155/2022/5315146. YixinQu, BingyingLin, ShuilingLi, et al. Effect of multichannel convolutional neural network-based model on the repair and aesthetic effect of eye plastic surgery patients[J]. Comput Math Methods Med, 2022, 2022: 5315146. DOI:10.1155/2022/5315146.
53、Varghese C, Harrison EM, O’Grady G, et al. Artificial intelligence in surgery[J]. Nat Med, 2024, 30(5): 1257-1268. DOI:10.1038/s41591-024-02970-3. Varghese C, Harrison EM, O’Grady G, et al. Artificial intelligence in surgery[J]. Nat Med, 2024, 30(5): 1257-1268. DOI:10.1038/s41591-024-02970-3.
54、van Leeuwen KG, Schalekamp S, Rutten MJCM, et al. Artificial intelligence in radiology: 100 commercially available products and their scientific evidence[J]. Eur Radiol, 2021, 31(6): 3797-3804. DOI:10.1007/s00330-021-07892-z.van Leeuwen KG, Schalekamp S, Rutten MJCM, et al. Artificial intelligence in radiology: 100 commercially available products and their scientific evidence[J]. Eur Radiol, 2021, 31(6): 3797-3804. DOI:10.1007/s00330-021-07892-z.
55、Alabi RO, Elmusrati M, Leivo I, et al. Artificial intelligence-driven radiomics in head and neck cancer: current status and future prospects[J]. Int J Med Inform, 2024, 188: 105464. DOI:10.1016/j.ijmedinf.2024.105464. Alabi RO, Elmusrati M, Leivo I, et al. Artificial intelligence-driven radiomics in head and neck cancer: current status and future prospects[J]. Int J Med Inform, 2024, 188: 105464. DOI:10.1016/j.ijmedinf.2024.105464.
56、Li Y, El Habib Daho M, Conze PH, et al. A review of deep learning-based information fusion techniques for multimodal medical image classification[J]. Comput Biol Med, 2024, 177: 108635. DOI:10.1016/j.compbiomed.2024.108635.Li Y, El Habib Daho M, Conze PH, et al. A review of deep learning-based information fusion techniques for multimodal medical image classification[J]. Comput Biol Med, 2024, 177: 108635. DOI:10.1016/j.compbiomed.2024.108635.
57、Kang D, Wu H, Yuan L, et al. A beginner’s guide to artificial intelligence for ophthalmologists[J]. Ophthalmol Ther, 2024, 13(7): 1841-1855. DOI:10.1007/s40123-024-00958-3. Kang D, Wu H, Yuan L, et al. A beginner’s guide to artificial intelligence for ophthalmologists[J]. Ophthalmol Ther, 2024, 13(7): 1841-1855. DOI:10.1007/s40123-024-00958-3.
58、Mohsen F, Ali H, El Hajj N, et al. Artificial intelligence-based methods for fusion of electronic health records and imaging data[J]. Sci Rep, 2022, 12(1): 17981. DOI:10.1038/s41598-022-22514-4. Mohsen F, Ali H, El Hajj N, et al. Artificial intelligence-based methods for fusion of electronic health records and imaging data[J]. Sci Rep, 2022, 12(1): 17981. DOI:10.1038/s41598-022-22514-4.
59、Wang L, Alexander CA. Big data analytics in medical engineering and healthcare: methods, advances and challenges[J]. J Med Eng Technol, 2020, 44(6): 267-283. DOI:10.1080/03091902.2020.1769758. Wang L, Alexander CA. Big data analytics in medical engineering and healthcare: methods, advances and challenges[J]. J Med Eng Technol, 2020, 44(6): 267-283. DOI:10.1080/03091902.2020.1769758.
60、Diwakar M, Singh P, Ravi V. Medical data analysis meets artificial intelligence (AI) and Internet of medical things (IoMT)[J]. Bioengineering, 2023, 10(12): 1370. DOI:10.3390/bioengineering10121370.Diwakar M, Singh P, Ravi V. Medical data analysis meets artificial intelligence (AI) and Internet of medical things (IoMT)[J]. Bioengineering, 2023, 10(12): 1370. DOI:10.3390/bioengineering10121370.
61、Manickam P, Mariappan SA, Murugesan SM, et al. Artificial intelligence (AI) and Internet of medical things (IoMT) assisted biomedical systems for intelligent healthcare[J]. Biosensors, 2022, 12(8): 562. DOI:10.3390/bios12080562.Manickam P, Mariappan SA, Murugesan SM, et al. Artificial intelligence (AI) and Internet of medical things (IoMT) assisted biomedical systems for intelligent healthcare[J]. Biosensors, 2022, 12(8): 562. DOI:10.3390/bios12080562.
62、Wu Y, Hu K, Chen DZ, et al. AI-enhanced virtual reality in medicine[C]//Proceedings of the Thirty-Third International Joint Conference on Artificial Intelligence. ACM, 2025: 8326-8334. DOI:10.24963/ijcai.2024/920.Wu Y, Hu K, Chen DZ, et al. AI-enhanced virtual reality in medicine[C]//Proceedings of the Thirty-Third International Joint Conference on Artificial Intelligence. ACM, 2025: 8326-8334. DOI:10.24963/ijcai.2024/920.
63、Moro%20C%2C%20%C5%A0tromberga%20Z%2C%20Raikos%20A%2C%20et%20al.%20The%20effectiveness%20of%20virtual%20and%20augmented%20reality%20in%20health%20sciences%20and%20medical%20anatomy%5BJ%5D.%20Anat%20Sci%20Educ%2C%202017%2C%2010(6)%3A%20549-559.%20DOI%3A10.1002%2Fase.1696.Moro%20C%2C%20%C5%A0tromberga%20Z%2C%20Raikos%20A%2C%20et%20al.%20The%20effectiveness%20of%20virtual%20and%20augmented%20reality%20in%20health%20sciences%20and%20medical%20anatomy%5BJ%5D.%20Anat%20Sci%20Educ%2C%202017%2C%2010(6)%3A%20549-559.%20DOI%3A10.1002%2Fase.1696.
64、Barteit%20S%2C%20Lanfermann%20L%2C%20B%C3%A4rnighausen%20T%2C%20et%20al.%20Augmented%2C%20mixed%2C%20and%20virtual%20reality-based%20head-mounted%20devices%20for%20medical%20education%3A%20systematic%20review%5BJ%5D.%20JMIR%20Serious%20Games%2C%202021%2C%209(3)%3A%20e29080.%20DOI%3A10.2196%2F29080.%20Barteit%20S%2C%20Lanfermann%20L%2C%20B%C3%A4rnighausen%20T%2C%20et%20al.%20Augmented%2C%20mixed%2C%20and%20virtual%20reality-based%20head-mounted%20devices%20for%20medical%20education%3A%20systematic%20review%5BJ%5D.%20JMIR%20Serious%20Games%2C%202021%2C%209(3)%3A%20e29080.%20DOI%3A10.2196%2F29080.%20
65、Fazlollahi AM, Bakhaidar M, Alsayegh A, et al. Effect of artificial intelligence tutoring vs expert instruction on learning simulated surgical skills among medical students: a randomized clinical trial[J]. JAMA Netw Open, 2022, 5(2): e2149008. DOI:10.1001/jamanetworkopen.2021.49008.Fazlollahi AM, Bakhaidar M, Alsayegh A, et al. Effect of artificial intelligence tutoring vs expert instruction on learning simulated surgical skills among medical students: a randomized clinical trial[J]. JAMA Netw Open, 2022, 5(2): e2149008. DOI:10.1001/jamanetworkopen.2021.49008.
66、Lu X, Zhou X, Song B, et al. Framework nucleic acids combined with 3D hybridization chain reaction amplifiers for monitoring multiple human tear cytokines[J]. Adv Mater, 2024, 36(26): e2400622. DOI:10.1002/adma.202400622.Lu X, Zhou X, Song B, et al. Framework nucleic acids combined with 3D hybridization chain reaction amplifiers for monitoring multiple human tear cytokines[J]. Adv Mater, 2024, 36(26): e2400622. DOI:10.1002/adma.202400622.
67、Irkham I, Ibrahim AU, Nwekwo CW, et al. Current technologies for detection of COVID-19: biosensors, artificial intelligence and Internet of medical things (IoMT): review[J]. Sensors, 2022, 23(1): 426. DOI:10.3390/s23010426. Irkham I, Ibrahim AU, Nwekwo CW, et al. Current technologies for detection of COVID-19: biosensors, artificial intelligence and Internet of medical things (IoMT): review[J]. Sensors, 2022, 23(1): 426. DOI:10.3390/s23010426.
68、Ghaffar Nia N, Kaplanoglu E, Nasab A. Evaluation of artificial intelligence techniques in disease diagnosis and prediction[J]. Discov Artif Intell, 2023, 3(1): 5. DOI:10.1007/s44163-023-00049-5.Ghaffar Nia N, Kaplanoglu E, Nasab A. Evaluation of artificial intelligence techniques in disease diagnosis and prediction[J]. Discov Artif Intell, 2023, 3(1): 5. DOI:10.1007/s44163-023-00049-5.
69、Badidi E. Edge AI for early detection of chronic diseases and the spread of infectious diseases: opportunities, challenges, and future directions[J]. Future Internet, 2023, 15(11): 370. DOI:10.3390/fi15110370.Badidi E. Edge AI for early detection of chronic diseases and the spread of infectious diseases: opportunities, challenges, and future directions[J]. Future Internet, 2023, 15(11): 370. DOI:10.3390/fi15110370.
70、Graves JS, Montalban X. Biosensors to monitor MS activity[J]. Mult Scler, 2020, 26(5): 605-608. DOI:10.1177/1352458519888178. Graves JS, Montalban X. Biosensors to monitor MS activity[J]. Mult Scler, 2020, 26(5): 605-608. DOI:10.1177/1352458519888178.
71、Zaslavsky J, Bannigan P, Allen C. Re-envisioning the design of nanomedicines: harnessing automation and artificial intelligence[J]. Expert Opin Drug Deliv, 2023, 20(2): 241-257. DOI:10.1080/17425247.2023.2167978. Zaslavsky J, Bannigan P, Allen C. Re-envisioning the design of nanomedicines: harnessing automation and artificial intelligence[J]. Expert Opin Drug Deliv, 2023, 20(2): 241-257. DOI:10.1080/17425247.2023.2167978.
72、Hayat H, Nukala A, Nyamira A, et al. A concise review: the synergy between artificial intelligence and biomedical nanomaterials that empowers nanomedicine[J]. Biomed Mater, 2021, 16(5): Biomedicalmaterials(Bristol+England)vol.16+510.1088/1748-605X/ac15b2.5Aug.2021+. DOI:10.1088/1748-605X/ac15b2.Hayat H, Nukala A, Nyamira A, et al. A concise review: the synergy between artificial intelligence and biomedical nanomaterials that empowers nanomedicine[J]. Biomed Mater, 2021, 16(5): Biomedicalmaterials(Bristol+England)vol.16+510.1088/1748-605X/ac15b2.5Aug.2021+. DOI:10.1088/1748-605X/ac15b2.
73、Cui Y, Wei Q, Park H, et al. Nanowire nanosensors for highly sensitive and selective detection of biological and chemical species[J]. Science, 2001, 293(5533): 1289-1292. DOI:10.1126/science.1062711.Cui Y, Wei Q, Park H, et al. Nanowire nanosensors for highly sensitive and selective detection of biological and chemical species[J]. Science, 2001, 293(5533): 1289-1292. DOI:10.1126/science.1062711.
74、Zheng G, Patolsky F, Cui Y, et al. Multiplexed electrical detection of cancer markers with nanowire sensor arrays[J]. Nat Biotechnol, 2005, 23(10): 1294-1301. DOI:10.1038/nbt1138.Zheng G, Patolsky F, Cui Y, et al. Multiplexed electrical detection of cancer markers with nanowire sensor arrays[J]. Nat Biotechnol, 2005, 23(10): 1294-1301. DOI:10.1038/nbt1138.
75、Lu Z, Liu T, Zhou X, et al. Rapid and quantitative detection of tear MMP-9 for dry eye patients using a novel silicon nanowire-based biosensor[J]. Biosens Bioelectron, 2022, 214: 114498. DOI:10.1016/j.bios.2022.114498. Lu Z, Liu T, Zhou X, et al. Rapid and quantitative detection of tear MMP-9 for dry eye patients using a novel silicon nanowire-based biosensor[J]. Biosens Bioelectron, 2022, 214: 114498. DOI:10.1016/j.bios.2022.114498.
76、Wang S, He X, Jian Z, et al. Advances and prospects of multi-modal ophthalmic artificial intelligence based on deep learning: a review[J]. Eye Vis, 2024, 11(1): 38. DOI:10.1186/s40662-024-00405-1. Wang S, He X, Jian Z, et al. Advances and prospects of multi-modal ophthalmic artificial intelligence based on deep learning: a review[J]. Eye Vis, 2024, 11(1): 38. DOI:10.1186/s40662-024-00405-1.
77、Li Z, Wang L, Wu X, et al. Artificial intelligence in ophthalmology: The path to the real-world clinic[J]. Cell Rep Med, 2023, 4(7): 101095. DOI:10.1016/j.xcrm.2023.101095. Li Z, Wang L, Wu X, et al. Artificial intelligence in ophthalmology: The path to the real-world clinic[J]. Cell Rep Med, 2023, 4(7): 101095. DOI:10.1016/j.xcrm.2023.101095.
78、Patel AU, Gu Q, Esper R, et al. The crucial role of interdisciplinary conferences in advancing explainable AI in healthcare[J]. BioMedInformatics, 2024, 4(2): 1363-1383. DOI:10.3390/biomedinformatics4020075. Patel AU, Gu Q, Esper R, et al. The crucial role of interdisciplinary conferences in advancing explainable AI in healthcare[J]. BioMedInformatics, 2024, 4(2): 1363-1383. DOI:10.3390/biomedinformatics4020075.
79、Tseng RMWW, Rim TH, Shantsila E, et al. Validation of a deep-learning-based retinal biomarker (Reti-CVD) in the prediction of cardiovascular disease: data from UK Biobank[J]. BMC Med, 2023, 21(1): 28. DOI:10.1186/s12916-022-02684-8.Tseng RMWW, Rim TH, Shantsila E, et al. Validation of a deep-learning-based retinal biomarker (Reti-CVD) in the prediction of cardiovascular disease: data from UK Biobank[J]. BMC Med, 2023, 21(1): 28. DOI:10.1186/s12916-022-02684-8.
80、Bertagnolli MM. Advancing health through artificial intelligence/machine learning: The critical importance of multidisciplinary collaboration[J]. PNAS Nexus, 2023, 2(12): pgad356. DOI:10.1093/pnasnexus/pgad356.Bertagnolli MM. Advancing health through artificial intelligence/machine learning: The critical importance of multidisciplinary collaboration[J]. PNAS Nexus, 2023, 2(12): pgad356. DOI:10.1093/pnasnexus/pgad356.
81、Stogiannos N, Gillan C, Precht H, et al. A multidisciplinary team and multiagency approach for AI implementation: a commentary for medical imaging and radiotherapy key stakeholders[J]. J Med Imaging Radiat Sci, 2024, 55(4): 101717. DOI:10.1016/j.jmir.2024.101717. Stogiannos N, Gillan C, Precht H, et al. A multidisciplinary team and multiagency approach for AI implementation: a commentary for medical imaging and radiotherapy key stakeholders[J]. J Med Imaging Radiat Sci, 2024, 55(4): 101717. DOI:10.1016/j.jmir.2024.101717.
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