Luo Yi, Hooshangnejad Hamed, Ngwa Wilfred, Ding Kai
Department of Biomedical Engineering, Johns Hopkins University, Baltimore, MD, USA.
Department of Radiation Oncology and Molecular Radiation Sciences, Johns Hopkins University, Baltimore, MD, USA.
Transl Lung Cancer Res. 2025 May 30;14(5):1830-1847. doi: 10.21037/tlcr-24-801. Epub 2025 May 23.
Lung cancer remains the leading cause of cancer-related deaths globally. Over the past decade, the development of artificial intelligence (AI) has significantly propelled lung cancer care, particularly in areas such as lung cancer early diagnosis, survival prediction, recurrence prediction, medical image processing, medical image registration, medical visual question answering, clinical report writing, medical image generation, and multimodal integration. This review aims to provide a comprehensive summary of the various AI methods utilized in lung cancer care, with a particular emphasis on machine learning and deep learning techniques. Moreover, with the advent and widespread application of large language models (LLMs), vision language models (VLMs), and multimodal integration for downstream clinical tasks, we explore the current landscape these cutting-edge AI tools offer. However, it also presents both significant challenges and opportunities, including data privacy risks, inherent biases that may exacerbate healthcare disparities, model hallucinations, ethical implications, implementation costs, and the lack of standardized evaluation metrics. Furthermore, the translation of these technologies from experimental research to clinical implementation demands comprehensive validation protocols and multidisciplinary collaboration to guarantee patient safety, therapeutic efficacy, and equitable healthcare delivery. This review emphasizes the critical role of AI in enhancing our understanding and management of lung cancer, ultimately striving for precision medicine and equitable healthcare worldwide.
肺癌仍然是全球癌症相关死亡的主要原因。在过去十年中，人工智能（AI）的发展显著推动了肺癌治疗，特别是在肺癌早期诊断、生存预测、复发预测、医学图像处理、医学图像配准、医学视觉问答、临床报告撰写、医学图像生成和多模态整合等领域。本综述旨在全面总结肺癌治疗中使用的各种人工智能方法，特别强调机器学习和深度学习技术。此外，随着大语言模型（LLMs）、视觉语言模型（VLMs）的出现及其在下游临床任务中的广泛应用以及多模态整合，我们探讨了这些前沿人工智能工具所呈现的现状。然而，它也带来了重大挑战和机遇，包括数据隐私风险、可能加剧医疗保健差距的固有偏差、模型幻觉、伦理影响、实施成本以及缺乏标准化评估指标。此外，将这些技术从实验研究转化为临床应用需要全面的验证方案和多学科合作，以确保患者安全、治疗效果和公平的医疗服务提供。本综述强调了人工智能在增强我们对肺癌的理解和管理方面的关键作用，最终致力于在全球范围内实现精准医疗和公平的医疗保健。
