大数据驱动的肿瘤病理智能分析：跨越形态学与多组学的诊疗桥梁

Big data-driven intelligent analysis of tumor pathology: a diagnostic and therapeutic bridge across morphology and multi-omics

摘要

系统总结了大数据与人工智能（artificial intelligence，AI）在肿瘤病理智能分析系统中的发展脉络、关键技术与临床转化应用。随着全切片成像（whole slide imaging，WSI）技术的成熟与普及，肿瘤病理分析正经历从传统数字病理（digital pathology，DP）向大数据驱动的AI计算病理学（computational pathology，CP）转型，为精准肿瘤学提供关键技术支撑。这一转型在技术层面体现为从早期的卷积神经网络（convolutional neural network，CNN）向具备全局视野的视觉transformer（vision transformer，ViT）及基础模型（foundation models，FMs）的架构升级，并利用弱监督学习（weakly supervised learning，WSL）与自监督学习（self-supervised learning，SSL）有效突破了高质量标注数据匮乏的瓶颈。在临床层面，AI已在肺癌、乳腺癌等全球高发癌种中广泛实践，不仅通过自动化分级解决了观察者间一致性难题，更展现出挖掘人眼无法识别的亚视觉特征的巨大潜力。多模态融合以及虚拟染色等生成式AI前沿技术的引入，正在进一步通过大同形态学与基因组学、转录组学的壁垒，为肿瘤诊断提供更全面的决策支撑。尽管目前仍需克服域偏移、可解释性及监管伦理等挑战，但建立人机协同的智能诊疗模式正成为提升诊断效率与实现精准医疗的关键演进方向，为跨越形态学与多组学的肿瘤诊疗搭建起核心技术桥梁。

Abstract

This paper systematically summarizes the development context, key technologies, and clinical translation applications of big data and artificial intelligence (AI) in intelligent analysis systems for tumor pathology. With the maturation and popularization of whole slide imaging (WSI) technology, tumor pathology analysis is undergoing a transformation from traditional digital pathology (DP) to big data-driven AI-based computational pathology (CP), providing critical technical support for precision oncology. Technically, this transformation is reflected in the architectural upgrade from early convolutional neural network (CNN) to vision transformer (ViT) with global receptive field and foundation models (FMs). Meanwhile, weakly supervised learning (WSL) and self-supervised learning (SSL) effectively break through the bottleneck of scarce high-quality annotated data. Clinically, AI has been widely applied in globally prevalent cancers such as lung cancer and breast cancer. It not only addresses inter-observer variability through automated grading but also demonstrates considerable potential in identifying subvisual features that are undetectable by human eyes. The introduction of cutting-edge generative AI technologies such as multimodal fusion and virtual staining is further breaking down the barriers between morpho-logy, genomics and transcriptomics, providing more comprehensive decision support for tumor diagnosis. Although challenges such as domain shift, interpretability, and regulatory and ethical concerns remain to be addressed, establishing a human-machine collaborative intelligent diagnosis and treatment model is becoming a key evolutionary direction for improving diagnostic efficiency and realizing precision medicine, building a core technical bridge for tumor diagnosis and treatment across morphology and multi-omics.

关键词

人工智能 / 临床病理 / 计算病理学 / 多组学融合 / 肿瘤精准诊疗 / 大数据 / 虚拟染色 / 智能诊疗模式

Key words

artificial intelligence / clinical pathology / computational pathology / multi-omics fusion / precision tumor diagnosis and treatment / big data / virtual staining / intelligent diagnosis and treatment model

参考文献

[1] 周桥. 肿瘤病理学研究的进展和展望[J]. 北京大学学报（医学版）, 2023, 55(2): 201-209.

[2] Zhou Q. Advances and prospects in oncologic pathology research[J]. Journal of Peking University（Medical Sciences）, 2023, 55(2): 201-209.

[3] Pantanowitz L, Sharma A, Carter A B, et al. Twenty years of digital pathology：an overview of the road travelled，what is on the horizon，and the emergence of vendor-neutral archives[J]. Journal of Pathology Informatics, 2018, 9(1): 40.

[4] Kumar N, Gupta R, Gupta S . Whole slide imaging（WSI）in pathology：current perspectives and future directions[J]. Journal of Digital Imaging, 2020, 33(4): 1034-1040.

[5] Hosseini M S, Bejnordi B E, Trinh V Q, et al. Computational pathology：a survey review and the way forward[J]. Journal of Pathology Informatics, 2024, 15: 100357.

[6] 姚家美, 张欣, 葛晓雯, 等. 数字病理切片在临床病理科住院医师规范化培训中的教学实践[J]. 中华病理学杂志, 2025, 54(10): 1114-1116.

[7] Yao J M, Zhang X, Ge X W, et al. Teaching practice of digital pathology slides in standardized residency training of clinical pathology[J]. Chinese Journal of Pathology, 2025, 54(10): 1114-1116.

[8] Walsh E, Orsi N M . The current troubled state of the global pathology workforce：a concise review[J]. Diagnostic Pathology, 2024, 19(1): 163.

[9] 袁姮, 霍欣燃, 姜文涛. 全息梯度差分卷积的图像分类网络[J]. 自动化学报, 2025, 51(9): 2106-2130.

[10] Yuan H, Huo X R, Jiang W T. Image classification network based on holographic gradient difference convolution[J]. Acta Automatica Sinica, 2025, 51(9): 2106-2130.

[11] Coudray N, Ocampo P S, Sakellaropoulos T, et al. Classification and mutation prediction from non-small cell lung cancer histopathology images using deep learning[J]. Nature medicine, 2018, 24(10): 1559-1567.

[12] Ehteshami Bejnordi B, Veta M, Johannes van Diest P, et al. Diagnostic assessment of deep learning algorithms for detection of lymph node metastases in women with breast cancer[J]. The Journal of the American Medical Association, 2017, 318(22): 2199-2210.

[13] Ebert N, Stricker D, Wasenmüller O . PLG-ViT：vision transformer with parallel local and global self-attention[J]. Sensors（Basel，Switzerland）, 2023, 23(7): 3447.

[14] Takahashi S, Sakaguchi Y, Kouno N, et al. Comparison of vision transformers and convolutional neural networks in medical image analysis：a systematic review[J]. Journal of Medical Systems, 2024, 48(1): 84.

[15] Yan R, Lv Z, Yang Z, et al. Sparse and hierarchical transformer for survival analysis on whole slide images[J]. IEEE Journal of Biomedical and Health Informatics, 2024, 28: 7-18.

[16] Li Z, Cong Y, Chen X, et al. Vision transformer-based weakly supervised histopathological image analysis of primary brain tumors[J]. iScience, 2023, 26(1): 105872.

[17] Zhou Y, Che S, Lu F, et al. Iterative multiple instance learning for weakly annotated whole slide image classification[J]. Physics in Medicine & Biology, 2023, 68(15): 155007.

[18] Lu M Y, Williamson D F K, Chen T Y, et al. Data-efficient and weakly supervised computational pathology on whole-slide images[J]. Nature Biomedical Engineering, 2021, 5(6): 555-570.

[19] Chanda T, Hauser K, Hobelsberger S, et al. Dermatologist-like explainable AI enhances trust and confidence in diagnosing melanoma[J]. Nature Communications, 2024, 15: 524.

[20] Vorontsov E, Bozkurt A, Casson A, et al. A foundation model for clinical-grade computational pathology and rare cancers detection[J]. Nature Medicine, 2024, 30(10): 2924-2935.

[21] Chen R J, Ding T, Lu M Y, et al. Towards a general-purpose foundation model for computational pathology[J]. Nature Medicine, 2024, 30(3): 850-862.

[22] Lu M Y, Chen B, Williamson D F K, et al. A visual-language foundation model for computational pathology[J]. Nature Medicine, 2024, 30(3): 863-874.

[23] Bray F, Laversanne M, Sung H, et al. Global cancer statistics，2022：GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries[J]. CA：A Cancer Journal for Clinicians, 2024, 74(3): 229-263.

[24] Kather J N, Pearson A T, Halama N, et al. Deep learning can predict microsatellite instability directly from histology in gastrointestinal cancer[J]. Nature Medicine, 2019, 25(7): 1054-1056.

[25] Bulten W, Kartasalo K, Chen P C, et al. Artificial intelligence for diagnosis and Gleason grading of prostate cancer：the PANDA challenge[J]. Nature Medicine, 2022, 28(1): 154-163.

[26] Song Z, Zou S, Zhou W, et al. Clinically applicable histopathological diagnosis system for gastric cancer detection using deep learning[J]. Nature communications, 2020, 11: 4294.

[27] Saillard C, Schmauch B, Laifa O, et al. Predicting survival after hepatocellular carcinoma resection using deep learning on histological slides[J]. Hepatology（Baltimore，Md）, 2020, 72(6): 2000-2013.

[28] Zhu X, Chen C, Guo Q, et al. Deep learning-based recognition of different thyroid cancer categories using whole frozen-slide images[J]. Frontiers in Bioengineering and Biotechnology, 2022, 10: 857377.

[29] Kanavati F, Hirose N, Ishii T, et al. A deep learning model for cervical cancer screening on liquid-based cytology specimens in whole slide images[J]. Cancers, 2022, 14(5): 1159.

[30] Harmon S A, Sanford T H, Brown G T, et al. Multiresolution application of artificial intelligence in digital pathology for prediction of positive lymph nodes from primary tumors in bladder cancer[J]. JCO clinical cancer informatics, 2020, 4: 367-382.

[31] Steinbuss G, Kriegsmann M, Zgorzelski C, et al. Deep learning for the classification of non-hodgkin lymphoma on histopathological images[J]. Cancers, 2021, 13(10): 2419.

[32] Bulten W, Balkenhol M, Belinga J A, et al. Artificial intelligence assistance significantly improves Gleason grading of prostate biopsies by pathologists[J]. Modern Pathology, 2021, 34(3): 660-671.

[33] Matsumoto H, Miyata R, Tsuruta Y, et al. Ki-67 evaluation using deep-learning model-assisted digital image analysis in breast cancer[J]. Histopathology, 2025, 86(3): 460-471.

[34] Zeng D T, Li M J, Lin R, et al. Prognostic role of Ki-67 in colorectal carcinoma：development and evaluation of machine learning prediction models[J]. World Journal of Clinical Oncology, 2025, 16(8): 107306.

[35] Park J, Shin S, Hwang W, et al. Deep learning predicts EGFR mutation status from histology images in non-small cell lung cancer[J]. Cancer Research Communications, 2025, 5(12): 2127-2141.

[36] Echle A, Grabsch H I, Quirke P, et al. Clinical-grade detection of microsatellite instability in colorectal tumors by deep learning[J]. Gastroenterology, 2020, 159(4): 1406-1416.e11.

[37] Xu Z, Li Y, Wang Y, et al. A deep learning quantified stroma-immune score to predict survival of patients with stage II-III colorectal cancer[J]. Cancer Cell International, 2021, 21(1): 585.

[38] 徐美东, 初元. 早期胃癌内镜诊治现状与挑战[J]. 中华消化杂志, 2025, 45(5): 309-313.

[39] Xu M D, Chu Y. Current status and challenges of endoscopic diagnosis and treatment of early gastric cancer[J]. Chinese Journal of Digestion, 2025, 45(5): 309-313.

[40] Liscia D S, D'Andrea M, Biletta E, et al. Use of digital pathology and artificial intelligence for the diagnosis of Helicobacter pylori in gastric biopsies[J]. Pathologica, 2022, 114(4): 295-303.

[41] Negrelli M, Frascarelli C, Maffini F, et al. Artificial intelligence in thyroid cytopathology：diagnostic and technical insights[J]. Cancers, 2025, 17(21): 3525.

[42] Ru Y, Tong X, Lin J, et al. Learning-based classification of B- and T-cell lymphoma on histopathological images：a multicenter study[J]. European Journal of Haematology, 2025, 115(2): 176-184.

[43] Song Y, Roy M, Zhong M, et al. Multimodal data fusion for whole-slide histopathology image classification[J]. Journal of Healthcare Informatics Research, 2025, 9(4): 513-532.

[44] Iacucci M, Santacroce G, Meseguer P, et al. Endo-Histo foundational fusion model：a novel artificial intelligence for assessing histologic remission and response to therapy in ulcerative colitis clinical trial[J]. Journal of Crohn's & Colitis, 2025, 19(7): jjaf108.

[45] Klöckner P, Teixeira J, Montezuma D, et al. H&E to IHC virtual staining methods in breast cancer：an overview and benchmarking[J]. NPJ Digital Medicine, 2025, 8: 384.

[46] Pati P, Karkampouna S, Bonollo F, et al. Accelerating histopathology workflows with generative AI-based virtually multiplexed tumour profiling[J]. Nature Machine Intelligence, 2024, 6(9): 1077-1093.

[47] Li Y, Pillar N, Liu T, et al. Label-free evaluation of lung and heart transplant biopsies using tissue autofluorescence-based virtual staining[J]. BME Frontiers, 2025, 6: 151.

[48] Chu ML, Ge XM, Eastham J, et al. Assessment of color reproducibility and mitigation of color variation in whole slide image scanners for toxicologic pathology[J]. Toxicologic Pathology, 2023, 51(6): 313-328.

[49] Michielli N, Caputo A, Scotto M, et al. Stain normalization in digital pathology：clinical multi-center evaluation of image quality[J]. Journal of Pathology Informatics, 2022, 13: 100145.

[50] Tellez D, Litjens G, Bándi P, et al. Quantifying the effects of data augmentation and stain color normalization in convolutional neural networks for computational pathology[J]. Medical Image Analysis, 2019, 58: 101544.

[51] Tosun A B, Pullara F, Becich M J, et al. Explainable AI（xAI）for anatomic pathology[J]. Advances in Anatomic Pathology, 2020, 27(4): 241-250.

[52] Klauschen F, Dippel J, Keyl P, et al. Toward explainable artificial intelligence for precision pathology[J]. Annual Review of Pathology：Mechanisms of Dsease, 2024, 19: 541-570.

[53] Matthews G A, McGenity C, Bansal D, et al. Public evidence on AI products for digital pathology[J]. NPJ Digital Mdicine, 2024, 7: 300.

[54] Xiong D D, He R Q, Huang Z G, et al. Global bibliometric mapping of the research trends in artificial intelligence-based digital pathology for lung cancer over the past two decades[J]. Journal of Medical Internet Research, 2024, 10: 20552076241277735.

基金资助

国家自然科学基金资助项目(NSFC82460783)

广西研究生教育创新计划项目(JGY2023068)

广西壮族自治区卫生健康委员会科研课题资助项目(Z-A20240554)