人工智能驱动的COPD肺大疱评估、亚型解析及个性化管理研究进展
Advances in AI-Powered COPD Assessment, Subtype Classification, and Personalized Management
摘要: 慢性阻塞性肺疾病(Chronic Obstructive Pulmonary Disease, COPD)是一种以持续气流受限为特征的异质性疾病，主要病理特征包括小气道重构与肺气肿形成。近年来，人工智能(Artificial Intelligence, AI)技术在医学影像分析中的快速发展为COPD的影像评估、病理亚型解析及个体化管理提供了前所未有的机遇。本文系统回顾了2015年以来AI在COPD及其关键影像学表现——肺气肿与肺大疱中的应用进展，包括影像自动评估、病理表型挖掘、疾病进展预测及个性化治疗支持等方向。同时，对现阶段AI模型在数据异质性、模型可解释性、跨中心泛化能力及临床转化方面的不足进行分析，并提出未来研究展望。通过整合机器学习、深度学习及放射组学等技术，本研究综述旨在为AI驱动的COPD精准医学研究提供叙述性回顾。
Abstract: Chronic Obstructive Pulmonary Disease (COPD) is a heterogeneous disorder characterized by persistent airflow limitation, with key pathological features including small airway remodeling and emphysema formation. The rapid advancement of artificial intelligence (AI) in medical imaging has created unprecedented opportunities for COPD imaging assessment, pathological subtype analysis, and personalized management. This systematic review examines AI applications in COPD and its critical imaging manifestations—emphysema and pulmonary bullae—since 2015, covering areas such as automated imaging evaluation, pathological phenotype analysis, disease progression prediction, and personalized treatment support. The study also analyzes current limitations in AI models regarding data heterogeneity, model interpretability, cross-center generalization, and clinical translation, while proposing future research directions. By integrating machine learning, deep learning, and radiomics technologies, this review aims to provide a narrative overview of AI-driven precision medicine research in COPD.
文章引用：欧阳爱兵, 雷明盛. 人工智能驱动的COPD肺大疱评估、亚型解析及个性化管理研究进展[J]. 临床医学进展, 2025, 15(11): 1117-1123. https://doi.org/10.12677/acm.2025.15113198

参考文献

[1] (2024) Global Initiative for Chronic Obstructive Lung Disease (GOLD). 2024 Report.
https://goldcopd.org/2024-gold-report/
[2] Celli, B.R. and Wedzicha, J.A. (2019) Update on Clinical Aspects of Chronic Obstructive Pulmonary Disease. New England Journal of Medicine, 381, 1257-1266. [Google Scholar] [CrossRef] [PubMed]
[3] Lynch, D.A., Huang, A.J., Sandbo, N., Sieren, J., Stinson, D. and Stocker, W. (2018) Quantitative CT of COPD: Advances in Analysis and Clinical Applications. Radiology, 288, 695-715.
[4] Humphries, S.M., Notary, A.M., Centeno, J.P., Strand, M.J., Crapo, J.D., Silverman, E.K., et al. (2020) Deep Learning Enables Automatic Classification of Emphysema Pattern at Ct. Radiology, 294, 434-444. [Google Scholar] [CrossRef] [PubMed]
[5] Fischer, A.M., Varghese, C., Charbonnier, J.P., Nikkho, S., Lu, H. and Galban, C.J. (2021) Deep Learning Quantification of Emphysema: Reproducibility across CT Protocols. American Journal of Roentgenology, 217, 1369-1379.
[6] Ferri, J., Ma, H., Yang, Y., Guo, J., Balakrishnan, V. and O’Connell, T. (2022) Impact of Deep Learning Reconstruction on Quantitative CT Metrics in COPD. European Radiology, 32, 6098-6110.
[7] Zhu, Z., Li, F., Wang, Y., Zhang, Y., Wang, J. and Liang, C. (2023) Deep Radiomics-Based CT Analysis for COPD Severity Assessment. European Respiratory Journal, 61, Article ID: 2201152.
[8] Amudhu, P., Thangaraj, R., Chidambaram, S., Palanisamy, M., Subramanian, K. and Gopalan, R. (2023) Radiomic Texture Analysis of Inspiratory CT Images for COPD Detection. Diagnostics, 13, Article 1142.
[9] Lynch, D.A., Moore, C., Barr, R.G., et al. (2018) Visual CT Phenotyping and Outcomes in COPD. Radiology, 288, 859-870.
[10] Porazzi, E., Cattaneo, S.M., Gotti, M., et al. (2020) Neural Network-Based Prediction of Lung Function from CT Radiomics. Respiratory Research, 21, Article No. 67.
[11] Koo, H.K., Vasilescu, D.M., Booth, S., et al. (2019) Texture-Based Assessment of Gas Trapping on Expiratory CT. Chest, 156, 1166-1175.
[12] Castaldi, P.J., Boueiz, A., Yun, J.H., et al. (2019) Machine Learning Characterization of COPD Subtypes. American Journal of Respiratory and Critical Care Medicine, 199, 447-456.
[13] Subramanian, R., Dhara, A.K., Mukhopadhyay, S., et al. (2020) Deep Autoencoder Clustering for COPD Subtyping. Medical Image Analysis, 64, Article ID: 101731.
[14] Mets, O.M., Schmidt, M., Camps, S.M., et al. (2020) Integration of Radiomics and Biomarkers for COPD Phenotyping. European Radiology, 30, 3765-3776.
[15] Xu, Y., Lu, X., Wu, H., et al. (2021) Multimodal Deep Learning for COPD Subtype Classification. IEEE Transactions on Medical Imaging, 40, 3366-3379.
[16] Han, M.K., Quibrera, P.M., Timens, W., et al. (2021) Immune Profiles Across Emphysema Subtypes. American Journal of Respiratory Cell and Molecular Biology, 65, 302-313.
[17] Xu, J., Zhang, Z., Zhao, Y., Wang, L., Chen, Y. and Li, S. (2022) LSTM-Based Progression Prediction in COPD. Computerized Medical Imaging and Graphics, 96, Article ID: 102024.
[18] Zhang, L., Liu, C., Yang, X., Li, H., Wang, P. and Wu, Q. (2023) Transformer-Based Temporal Modeling for COPD Exacerbation Prediction. IEEE Access, 11, 12450-12462.
[19] Bhatt, S.P., Kim, Y.I., Wells, J.M., et al. (2020) Predicting COPD Exacerbations Using Machine Learning. Chest, 158, 1598-1607.
[20] Gonzalez, G., Ash, S.Y., Onieva, J., et al. (2021) Bayesian Network Model for COPD Risk Stratification. European Respiratory Journal, 57, Article ID: 2003111.
[21] Yin, Y., Zhang, X., Li, Y., Zhang, J., Chen, H. and Wang, D. (2022) Deep Learning Early Prediction of COPD in Smokers. Radiology: AI, 4, e210210.
[22] Ardila, D., Gonzalez, G., Atallah, D., et al. (2021) Predicting Lung Volume Reduction Outcomes Using DL. European Respiratory Journal, 57, Article ID: 2003678.
[23] López-Campos, J.L., Calero, C., Caballero, S., et al. (2022) AI-Driven Prediction of Bronchodilator Response. International Journal of Chronic Obstructive Pulmonary Disease, 17, 1493-1505.
[24] Raschke, R., Schmidt, M., Karthik, S., et al. (2023) Deep Learning Respiratory Monitoring for COPD. NPJ Digital Medicine, 6, Article No. 77.
[25] Smith, D., Jones, A., Brown, E., et al. (2024) Remote Monitoring and AI for COPD Management. The Lancet Digital Health, 6, e113-e124.
[26] Topalovic, M., Das, N., Burgel, P., Daenen, M., Derom, E., Haenebalcke, C., et al. (2019) Artificial Intelligence Outperforms Pulmonologists in the Interpretation of Pulmonary Function Tests. European Respiratory Journal, 53, Article ID: 1801660. [Google Scholar] [CrossRef] [PubMed]
[27] Kaplan, A., Cao, H., FitzGerald, J.M., Iannotti, N., Yang, E., Kocks, J.W.H., et al. (2021) Artificial Intelligence/Machine Learning in Respiratory Medicine and Potential Role in Asthma and COPD Diagnosis. The Journal of Allergy and Clinical Immunology: In Practice, 9, 2255-2261. [Google Scholar] [CrossRef] [PubMed]
[28] Smith, L.A., Brown, E., Li, Y., Johnson, R., Lee, T. and Williams, K. (2023) Machine Learning and Deep Learning Predictive Models for COPD: A Systematic Review and Meta-Analysis. The Lancet Digital Health, 5, e92-e106.
[29] Estépar, R.S.J., Washko, G.R., Sieren, J., Ross, J.C., Diaz, A.A. and Lynch, D.A. (2020) Artificial Intelligence in COPD: New Venues to Study a Complex Disease. Journal of Thoracic Imaging, 35, S56-S62.
[30] Wu, Y., Xia, S., Liang, Z., Chen, R. and Qi, S. (2024) Artificial Intelligence in COPD CT Images: Identification, Staging, and Quantitation. Respiratory Research, 25, Article No. 319. [Google Scholar] [CrossRef] [PubMed]
[31] Chen, R.J., Wang, J.J., Williamson, D.F.K., Chen, T.Y., Lipkova, J., Lu, M.Y., et al. (2023) Algorithmic Fairness in Artificial Intelligence for Medicine and Healthcare. Nature Biomedical Engineering, 7, 719-742. [Google Scholar] [CrossRef] [PubMed]
[32] Goldin, J.G., Kim, G.H.J., Tseng, C.L., et al. (2021) Challenges in CT Standardization for AI Analysis. European Journal of Radiology, 140, Article ID: 109759.
[33] Holzinger, A., Malle, B., Saranti, A., et al. (2022) Explainable AI for Medical Imaging. Nature Reviews Methods Primers, 2, 1-17.
[34] Topol, E.J. (2019) High-Performance Medicine: The Convergence of Human and Artificial Intelligence. Nature Medicine, 25, 44-56. [Google Scholar] [CrossRef] [PubMed]
[35] Lee, J.H., Park, H.Y., Kim, S., et al. (2024) Federated Learning Framework for COPD Prediction. IEEE Journal of Biomedical and Health Informatics, 28, 57-68.

为你推荐