人工智能融合临床与多组学数据在卒中防治及医药研发中的应用与挑战

doi:10.3969/j.issn.1673-5765.2025.06.006

中国卒中杂志 ›› 2025, Vol. 20 ›› Issue (6): 710-717.DOI: 10.3969/j.issn.1673-5765.2025.06.006

人工智能融合临床与多组学数据在卒中防治及医药研发中的应用与挑战

勾岚¹，姜明慧^1，2，姜勇³，廖晓凌⁴，李昊^1，2，5，张杰^1，2，程丝^1，2，5

1 北京 100070 首都医科大学附属北京天坛医院，国家神经系统疾病临床医学研究中心，卒中多组学创新中心
2 脑血管病药械研发北京市重点实验室
3 首都医科大学附属北京天坛医院，国家神经系统疾病临床医学研究中心
4 首都医科大学附属北京天坛医院神经病学中心
5 首都医科大学卒中精准临床诊疗与研究中心

收稿日期:2025-03-28 出版日期:2025-06-20 发布日期:2025-06-20
通讯作者: 张杰 zhangjie@ncrcnd.org.cn 程丝 sicheng@ncrcnd.org.cn
基金资助:
国家重点研发计划（2022YFE0209600）
国家自然科学基金（82471304）
中国科协青年人才托举工程（2023QNRC001）

Applications and Challenges of Integrating Artificial Intelligence with Clinical and Multi-omics Data in Stroke Prevention, Treatment, and Pharmaceutical Research and Development

GOU Lan¹, JIANG Minghui^1,2, JIANG Yong³, LIAO Xiaoling⁴, LI Hao^1,2,5, ZHANG Jie^1,2, CHENG Si^1,2,5

1 Center of Excellence for Omics Research, China National Clinical Research Center for Neurological Diseases, Beijing Tiantan Hospital, Capital Medical University, Beijing 100070, China
2 Beijing Key Laboratory of Drug and Device Research and Development for Cerebrovascular Diseases, Beijing 100070, China
3 China National Clinical Research Center for Neurological Diseases, Beijing 100070, China
4 Department of Neurology, Beijing Tiantan Hospital, Capital Medical University, Beijing 100070, China
5 Clinical Center for Precision Medicine in Stroke, Capital Medical University, Beijing 100070, China

Received:2025-03-28 Online:2025-06-20 Published:2025-06-20
Contact: ZHANG Jie, E-mail: zhangjie@ncrcnd.org.cn CHENG Si, E-mail: sicheng@ncrcnd.org.cn

摘要/Abstract

摘要： 卒中是全球致死致残的首要病因。基于人工智能（artificial intelligence，AI）融合临床与多组学数据的研究范式，为卒中精准防治提供了新方案。AI技术通过整合分析临床与多组学数据，能够提升卒中高危人群识别能力，优化早期诊断与风险评估，实现卒中亚型的精准分型，同时可筛选潜在药物靶点，构建预后预测模型。然而，卒中多组学资源相对匮乏、多模态数据融合困难，以及算法可解释性不足等仍是临床转化的主要瓶颈。本文对AI融合临床与多组学数据在卒中防治及医药研发中的应用与挑战等研究进展进行综述。

文章导读： 人工智能融合临床与多组学数据，显著提升了卒中预测、诊断、治疗的精准性与药物研发进度。

关键词: 卒中防治; 人工智能; 多组学; 精准医疗; 医药研发

Abstract: Stroke is the leading global cause of death and disability. The research paradigm that integrates artificial intelligence (AI) with clinical and multi-omics data provides novel solutions for the precision prevention and treatment of stroke. By integrating and analyzing clinical and multi-omics data, AI technology enhances the identification of high-risk populations, optimizes early diagnosis and risk assessment, enables precise subtyping of stroke, facilitates the screening of potential drug targets, and constructs prognostic prediction models. However, critical challenges, such as insufficient multi-omics resources, difficulties in multi modal data integration, and limited interpretability of algorithms, remain major bottlenecks in clinical translation. This article reviews the advances and challenges in AI-driven integration of clinical and multi-omics data in stroke prevention and treatment as well as pharmaceutical research and development.

Key words: Stroke prevention and treatment; Artificial intelligence; Multi-omics; Precision medicine; Pharmaceutical research and development

中图分类号:

勾岚, 姜明慧, 姜勇, 廖晓凌, 李昊, 张杰, 程丝. 人工智能融合临床与多组学数据在卒中防治及医药研发中的应用与挑战[J]. 中国卒中杂志, 2025, 20(6): 710-717.

GOU Lan, JIANG Minghui, JIANG Yong, LIAO Xiaoling, LI Hao, ZHANG Jie, CHENG Si. Applications and Challenges of Integrating Artificial Intelligence with Clinical and Multi-omics Data in Stroke Prevention, Treatment, and Pharmaceutical Research and Development[J]. Chinese Journal of Stroke, 2025, 20(6): 710-717.

参考文献

[1] GBD 2021 Stroke Risk Factor Collaborators. Global，regional，and national burden of stroke and its risk factors，1990—2021：a systematic analysis for the global burden of disease study 2021[J]. Lancet Neurol，2024，23（10）：973-1003.
[2] 《中国脑卒中防治报告2021》编写组. 《中国脑卒中防治报告2021》概要[J]. 中国脑血管病杂志，2023，20（11）：783-793.
Report on Stroke Prevention and Treatment in China Writing Group. Brief report on stroke prevention and treatment in China，2021[J]. Chin J Cerebrovasc Dis，2023，20（11）：783-793.
[3] 刘斯洋，林星辰，程丝，等. 多组学大数据与医学发展[J]. 科技导报，2024，42（12）：51-74.
LIU S Y，LIN X C，CHENG S，et al. Multi-omics big data and medical advancements[J]. Sci Technol Rev，2024，42（12）：51-74.
[4] GOODWIN S，MCPHERSON J D，MCCOMBIE W R. Coming of age：ten years of next-generation sequencing technologies[J]. Nat Rev Genet，2016，17（6）：333-351.
[5] ALONSO-BETANZOS A，BOLÓN-CANEDO V. Big-data analysis，cluster analysis，and machine-learning approaches[J/OL]. Adv Exp Med Biol，2018，1065：607-626[2025-02-19]. https://doi.org/10.1007/978-3-319-77932-4_37.
[6] CUI C，YANG H C，WANG Y H，et al. Deep multimodal
fusion of image and non-image data in disease diagnosis and prognosis：a review[J/OL]. Prog Biomed Eng（Bristol），2023，5（2）：10.1088/2516-1091/acc2fe[2025-02-19]. https://doi.org/10.1088/2516-1091/acc2fe.
[7] KANG M，KO E，MERSHA T B. A roadmap for multi-omics data integration using deep learning[J/OL]. Brief Bioinform，2022，23（1）：bbab454[2025-02-19]. https://doi.org/10.1093/bib/bbab454.
[8] LIPKOVA J，CHEN R J，CHEN B W，et al. Artificial intelligence for multimodal data integration in oncology[J]. Cancer Cell，2022，40（10）：1095-1110.
[9] SHAFAAT O，BERNSTOCK J D，SHAFAAT A，et al. Leveraging artificial intelligence in ischemic stroke imaging[J]. J Neuroradiol，2022，49（4）：343-351.
[10] QUISPE-OROZCO D，FAROOQUI M，ZEVALLOS C，et al. Angiography suite cone-beam computed tomography perfusion imaging in large-vessel occlusion patients using RAPID software：a pilot study[J/OL]. Stroke，2021，52（9）：e542-e544[2025-02-19]. https://doi.org/10.1161/STROKEAHA.121.035992.
[11] BOUTS M J R J，TIEBOSCH I A C W，VAN DER TOORN A，et al. Early identification of potentially salvageable tissue with MRI-based predictive algorithms after experimental ischemic stroke[J]. J Cereb Blood Flow Metab，2013，33（7）：1075-1082.
[12] RAJPURKAR P，LUNGREN M P. The current and future state of AI interpretation of medical images[J]. N Engl J Med，2023，388（21）：1981-1990.
[13] LI W T，SHAO C Y，ZHOU H F，et al. Multi-omics research strategies in ischemic stroke：a multidimensional perspective[J/OL]. Ageing Res Rev，2022，81：101730[2025-02-19]. https://doi.org/10.1016/j.arr.2022.101730.
[14] VAN SCHALKWIJK D B，DE GRAAF A A，TSIVTSIVADZE E，et al. Lipoprotein metabolism indicators improve cardiovascular risk prediction[J/OL]. PLoS One，2014，9（3）：e92840[2025-02-19].
https://doi.org/10.1371/journal.pone.0092840.
[15] FAN Y S，SONG Z S，ZHANG M Q. Emerging frontiers of artificial intelligence and machine learning in ischemic stroke：a comprehensive investigation of state-of-the-art methodologies，clinical applications，and unraveling challenges[J]. EPMA J，2023，14（4）：645-661.
[16] NTAIOS G，BAUMGARTNER H，DOEHNER W，et al. Embolic strokes of undetermined source：a clinical consensus statement of the ESC Council on Stroke，the European Association of Cardiovascular Imaging and the European Heart Rhythm Association of the ESC[J]. Eur Heart J，2024，45（19）：1701-1715.
[17] KLEINDORFER D O，TOWFIGHI A，CHATURVEDI S，et al. 2021 guideline for the prevention of stroke in patients with stroke and transient ischemic attack：a guideline from the American Heart Association/American Stroke Association[J/OL]. Stroke，52（7），e364-e467[2025-02-19]. https://doi.org/10.1161/STR.0000000000000375.
[18] DING L L，LIU C，LI Z X，et al. Incorporating artificial intelligence into stroke care and research[J/OL]. Stroke，2020，51（12）：e351-e354[2025-02-19]. https://doi.org/10.1161/STROKEAHA.120.031295.
[19] HOU Y C C，YU H C，MARTIN R，et al. Precision medicine integrating whole-genome sequencing，comprehensive metabolomics，and advanced imaging[J]. Proc Natl Acad Sci U S A，2020，117（6）：3053-3062.
[20] LIU Y，WANG W L，CUI X，et al. Exploring genetic associations of 3 types of risk factors with ischemic stroke：an integrated bioinformatics study[J]. Stroke，2024，55（6）：1619-1628.
[21] BARAGETTI A，GRIGORE L，OLMASTRONI E，et al. Plasma proteins associate with carotid plaques and predict incident atherosclerotic cardiovascular events[J/OL]. Vascul Pharmacol，2024，156：107394[2025-02-19]. https://doi.org/10.1016/j.vph.2024.107394.
[22] THEOFILATOS K，STOJKOVIC S，HASMAN M，et al. Proteomic atlas of atherosclerosis：the contribution of proteoglycans to sex differences，plaque phenotypes，and outcomes[J]. Circ Res，2023，133（7）：542-558.
[23] ZHANG F，ZHANG Y，ZHOU Q，et al. Using machine learning to identify proteomic and metabolomic signatures of stroke in atrial fibrillation[J/OL]. Comput Biol Med，2024，173：108375[2025-02-19]. https://doi.org/10.1016/j.compbiomed.2024.108375.
[24] SONG J W，ZAIDI S A A，HE L G，et al. Integrative analysis of machine learning and molecule docking simulations for ischemic stroke diagnosis and therapy[J/OL]. Molecules，2023，28（23）：7704[2025-02-19]. https://doi.org/10.3390/molecules28237704.
[25] FAN J X，CHEN M Y，CAO S，et al. Identification of a ferroptosis-related gene pair biomarker with immune infiltration landscapes in ischemic stroke：a bioinformatics-based comprehensive study[J/OL]. BMC Genomics，2022，23（1）：59[2025-02-19]. https://doi.org/10.1186/s12864-022-08295-0.
[26] QIN X H，YI S F，RONG J T，et al. Identification of anoikis-related genes classification patterns and immune infiltration characterization in ischemic stroke based on machine learning[J/OL]. Front Aging Neurosci，2023，15：1142163[2025-02-19]. https://doi.org/10.3389/fnagi.2023.1142163.
[27] REN P，WANG J Y，CHEN H L，et al. Diagnostic model constructed by nine inflammation-related genes for diagnosing ischemic stroke and reflecting the condition of immune-related cells[J/OL]. Front Immunol，2022，13：1046966[2025-02-19]. https://doi.org/10.3389/fimmu.2022.1046966.
[28] ZOMPOLA A，KORFIATI A，THEOFILATOS K，et al. Omics-CNN：a comprehensive pipeline for predictive analytics in quantitative omics using one-dimensional convolutional neural networks[J/OL]. Heliyon，2023，9（11）：e21165[2025-02-19]. https://doi.org/10.1016/j.heliyon.2023.e21165.
[29] TIEDT S，BRANDMAIER S，KOLLMEIER H，et al. Circulating metabolites differentiate acute ischemic stroke from stroke mimics[J]. Ann Neurol，2020，88（4）：736-746.
[30] LEE H J，SCHWAMM L H，SANSING L H，et al. StrokeClassifier：ischemic stroke etiology classification by ensemble consensus modeling using electronic health records[J/OL]. NPJ Digit Med，2024，7（1）：130[2025-02-19]. https://doi.org/10.1038/s41746-024-01120-w.
[31] BANG J H，KIM E H，KIM H J，et al. Machine learning-based etiologic subtyping of ischemic stroke using circulating exosomal microRNAs[J/OL]. Int J Mol Sci，2024，25（12）：6761[2025-02-19]. https://doi.org/10.3390/ijms25126761.
[32] LI W，BAI X S，HAO J H，et al. Thrombosis origin identification of cardioembolism and large artery atherosclerosis by distinct metabolites[J]. J Neurointerv Surg，2023，15（7）：701-707.
[33] ZHANG J X，ZHANG B B，LI T T，et al. Exploring the shared biomarkers between cardioembolic stroke and atrial fibrillation by WGCNA and machine learning[J/OL]. Front Cardiovasc Med，2024，11：1375768
[2025-02-19]. https://doi.org/10.3389/fcvm.2024.1375768.
[34] WANG X，MENG X Y，MENG L B，et al. Joint efficacy of the three biomarkers SNCA，GYPB and HBG1 for atrial fibrillation and stroke：analysis via the support vector machine neural network[J]. J Cell Mol Med，2022，26（7）：2010-2022.
[35] DING L L，LIU Y，MENG X，et al. Biomarker and genomic analyses reveal molecular signatures of non-cardioembolic ischemic stroke[J/OL]. Signal Transduct Target Ther，2023，8（1）：222[2025-02-19]. https://doi.org/10.1038/s41392-023-01465-w.
[36] MARTHA S R，LEVY S H，FEDERICO E，et al. Machine learning analysis of the cerebrovascular thrombi lipidome in acute ischemic stroke[J]. J Neurosci Nurs，2023，55（1）：10-17.
[37] AMENT Z，BEVERS M B，WOLCOTT Z，et al. Uric acid and gluconic acid as predictors of hyperglycemia and cytotoxic injury after stroke[J]. Transl Stroke Res，2021，12（2）：293-302.
[38] ZHAO Y W，MA X Y，MENG X H，et al. Integrating machine learning and single-cell transcriptomic analysis to identify potential biomarkers and analyze immune features of ischemic stroke[J/OL]. Sci Rep，2024，14（1）：26069[2025-02-19]. https://doi.org/10.1038/s41598-024-77495-3.
[39] HUANG D，ZHU Y，SHEN J F，et al. Identification of potential neddylation-related key genes in ischemic stroke based on machine learning methods[J]. Mol Neurobiol，2024，61（5）：2530-2541.
[40] YU H Y，JI X Y，OUYANG Y. Unfolded protein response pathways in stroke patients：a comprehensive landscape assessed through machine learning algorithms and experimental verification[J/OL]. J Transl Med，2023，21（1）：759[2025-02-19]. https://doi.org/10.1186/s12967-023-04567-9.
[41] LUO J，CAI Y，XIAO P，et al. Inflammation-derived and clinical indicator-based predictive model for ischemic stroke recovery[J/OL]. J Am Heart Assoc，2024，13（15）：e035609[2025-02-19]. https://doi.org/10.1161/JAHA.124.035609.
[42] CHI N F，CHANG T H，LEE C Y，et al. Untargeted metabolomics predicts the functional outcome of ischemic stroke[J]. J Formos Med Assoc，2021，120（1 Pt 1）：234-241.
[43] CHEN S D，XU Z，YIN J F，et al. Predicting functional outcome in ischemic stroke patients using genetic，environmental，and clinical factors：a machine learning analysis of population-based prospective cohort study[J/OL]. Brief Bioinform，2024，25（6）：bbae487[2025-02-19]. https://doi.org/10.1093/bib/bbae487.
[44] SHAMEER K，JOHNSON K W，GLICKSBERG B S，et al. Machine learning in cardiovascular medicine：are we there yet？[J]. Heart，2018，104（14）：1156-1164.
[45] RATWANI R M，BENDA N C，HETTINGER A Z，et al. Electronic health record vendor adherence to usability certification requirements and testing standards[J]. JAMA，2015，314（10）：1070-1071.
[46] MULLARD A. The UK biobank at 20[J]. Nat Rev Drug Discov，2022，21（9）：628-629.
[47] DEVINENI D，AKBARPOUR M，GONG Y F，et al. Inadequate use of newer treatments and glycemic control by cardiovascular risk and sociodemographic groups in US adults with diabetes in the NIH precision medicine initiative all of us research program[J]. Cardiovasc Drugs Ther，2024，38（2）：347-357.
[48] REEL P S，REEL S，PEARSON E，et al. Using machine learning approaches for multi-omics data analysis：a review[J/OL]. Biotechnol Adv，2021，49：107739[2025-02-19]. https://doi.org/10.1016/j.biotechadv.2021.107739.
[49] TANVIR R B，ISLAM M M，SOBHAN M，et al. MOGAT：a multi-omics integration framework using graph attention networks for cancer subtype prediction[J/OL]. Int J Mol Sci，2024，25（5）：2788[2025-02-19]. https://doi.org/10.3390/ijms25052788.
[50] WANG X H，YE S S，FENG J W，et al. Performance of ChatGPT on prehospital acute ischemic stroke and large vessel occlusion（LVO）stroke screening[J/OL]. Digit Health，2024，10：20552076241297127[2025-02-19]. https://doi.org/10.1177/20552076241297127.
[51] LEHNEN N C，DORN F，WIEST I C，et al. Data extraction from free-text reports on mechanical thrombectomy in acute ischemic stroke using ChatGPT：a retrospective analysis[J/OL]. Radiology，2024，311（1）：e232741[2025-02-19]. https://doi.org/10.1148/radiol.232741.
[52] CHEN T C，COULDWELL M W，SINGER J，et al. Assessing the clinical reasoning of ChatGPT for mechanical thrombectomy in patients with stroke[J]. J Neurointerv Surg，2024，16（3）：253-260.
[53] 东软医疗：DeepSeek强势来袭，AI赋能医疗新高度[EB/OL].（2025-02-10）[2025-02-19]. http://www.zhonghongwang.com/show-140-371405-1.html.
[54] KLAUSCHEN F，DIPPEL J，KEYL P，et al. Toward explainable artificial intelligence for precision pathology[J/OL]. Annu Rev Pathol，2024，19：541-570[2025-02-19]. https://doi.org/10.1146/annurev-pathmechdis-051222-113147.
[55] AN N A，ZHANG J，MO F，et al. De novo genes with an lncRNA origin encode unique human brain developmental functionality[J]. Nat Ecol Evol，2023，7（2）：264-278.
[56] KÖNIG H，FRANK D，BAUMANN M，et al. AI models and the future of genomic research and medicine：true sons of knowledge？[J/OL]. Bioessays，2021，43（10）：e2100025[2025-02-19]. https://doi.org/10.1002/bies.202100025.
[57] HENRY K E，KORNFIELD R，SRIDHARAN A，et al. Human-machine teaming is key to AI adoption：clinicians’ experiences with a deployed machine learning system[J/OL]. NPJ Digit Med，2022，5（1）：97[2025-02-19]. https://doi.org/10.1038/s41746-022-00597-7.

人工智能融合临床与多组学数据在卒中防治及医药研发中的应用与挑战

Applications and Challenges of Integrating Artificial Intelligence with Clinical and Multi-omics Data in Stroke Prevention, Treatment, and Pharmaceutical Research and Development

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

[1]	张杰, 李昊, 程丝. 脑血管病的多组学研究[J]. 中国卒中杂志, 2025, 20(6): 657-662.
[2]	许喆, 石延枫, 张杰, 姜明慧, 刘阳, 李昊, 廖晓凌, 程丝. 全基因组关联分析标准化流程的构建与扩展应用[J]. 中国卒中杂志, 2025, 20(6): 664-674.
[3]	周宏宇, 李子孝. 数字生物标志物：打开数智医疗应用的钥匙[J]. 中国卒中杂志, 2025, 20(4): 385-390.
[4]	池琦, 赵颖, 韩东倩, 张锶琪, 杜沛洁, 董琬玥, 徐安定, 杨振国, 孟珩. 急性缺血性卒中发病时间评估的研究进展[J]. 中国卒中杂志, 2025, 20(4): 493-499.
[5]	樊燕琴, 曹黎明, 任力杰. 我国社区卒中防治：医防融合视角下的卒中防治现状、挑战与策略分析[J]. 中国卒中杂志, 2025, 20(3): 261-268.
[6]	常丹凤, 申斗, 曹欣, 马可菲, 杨燕绥, 张玉哲, 池枫, 刘跃华, 高月霞. 基于社区的卒中全生命周期健康管理模式研究进展[J]. 中国卒中杂志, 2025, 20(3): 298-304.
[7]	张淋源, 沈翔, 吕海燕, 王国栋, 吴云成. 人工智能在提升神经科住院医师脑血管病诊治能力中的应用与挑战[J]. 中国卒中杂志, 2025, 20(3): 380-383.
[8]	邳靖陶, 尹伟, 宋晓微, 武海岩, 梁怡凡, 高策舒, 魏宸铭, 桑振华, 安志明, 江燕敏, 吴雅婷, 牟婧宇, 赵心怡, 陈乐, 武剑. 不同年资神经内科医师对数智化诊疗的认知、态度及功能需求现状调查[J]. 中国卒中杂志, 2025, 20(1): 78-86.
[9]	吴春艳, 尹雅诗, 王广志, 岳奎涛. 急性缺血性卒中不同时间窗影像学评价及应用进展[J]. 中国卒中杂志, 2024, 19(9): 1094-1101.
[10]	孟令涉, 王春娟. 人工智能与机器学习在心脑血管疾病管理中的应用与前景：美国心脏学会使用人工智能改善心脏疾病结局科学声明解读[J]. 中国卒中杂志, 2024, 19(6): 621-631.
[11]	王秀杰, 徐安定. 中国万人卒中多组学多模态数据库（STROMICS）：提升卒中精准诊疗的宝贵资源[J]. 中国卒中杂志, 2024, 19(12): 1369-1371.
[12]	杨溢, 刘清源, 刘伟奇, 王硕. 基于人工智能的未破裂颅内动脉瘤形态学指标三维测量方法[J]. 中国卒中杂志, 2024, 19(1): 112-119.
[13]	张心邈, 徐曼, 丁玲玲, 荆京, 龚浠平, 董可辉, 赵性泉, 王拥军, 李子孝. 脑血管病临床决策支持系统对卒中医疗服务质量关键绩效指标的影响研究[J]. 中国卒中杂志, 2024, 19(1): 120-124.
[14]	梁宝生, 周江杰, 王胜锋. 基于贝叶斯加性回归树的个体化治疗效应估计方法及应用[J]. 中国卒中杂志, 2023, 18(07): 770-779.
[15]	陈悦薇, 徐群. 脑小血管病的外周标志物及其多组学研究进展[J]. 中国卒中杂志, 2023, 18(03): 340-346.