中国肺癌的发病率及死亡率位居世界首位，肺癌的早期发现和早期诊断是肺癌防治、提高生存率的关键。肺腺癌侵袭性的精准评估直接决定了治疗方案的选择和预后评估。依靠传统影像学、计算机辅助诊断及影像组学评估肺腺癌侵袭性，其准确性、鲁棒性均受到一定限制。深度学习方法为实现医学图像的自动深度分析及高精度智能辅助诊断提供了新的契机。目前诊断效能的提高面临两大瓶颈性问题：一是扫描方案和重建方式标准尚未建立；二是传统CT图像的分辨率无法支持对肿瘤及周围组织的相关性开展深入研究。我们前期研究表明瘤周影像组学信息及先验领域知识可以提高深度学习模型输出效能。为此，本研究基于体模和人体数据集建立标准化扫描方案，在此基础上通过创新的靶重建和图像映射技术重建超分辨率影像，最终建立融合瘤周信息及先验领域知识的肺腺癌侵袭性预警深度学习模型，以期实现早期肺腺癌智能识别及侵袭性预判，为外科手术治疗方案选择及预后评估提供依据。

The incidence and mortality of lung cancer in China ranks first in the world. Early detection and early diagnosis of lung cancer have become the primary methods of prevention and treatment of lung cancer. Precise assessment of invasiveness for lung adenocarcinoma directly determines the choice of treatment and prognosis. The accuracy, robustness and reliability of traditional CT signs, computer-aided and imaging methods are all subject to certain restrictions. The deep learning methods provide a new opportunity to realize the automatic analysis of medical images and intelligent auxiliary diagnosis with high accuracy. At present, to improve the diagnostic efficiency of deep learning is confronted with two major bottlenecks: on the one hand, there is no unified standard for scanning scheme and reconstruction method; on the other hand, the present studies usually ignore the image characteristics of the area around the tumour. Our previous study has proved that the characteristics of radiomics around the tumour and prior domain knowledge could improve the output efficiency of the deep learning model. Therefore, this study intends to standardize scanning schemes based on the human model and human dataset. On this foundation, to reconstruct the super-resolution image through the method of target reconstructing and mapping technique of image, and, finally establish the deep learning model of lung invasiveness with early warning function combined with peritumoral imaging features and prior domain knowledge. We hope this study could be helpful for intelligent recognition and invasive prediction of lung adenocarcinoma and provide the basis for the choice of surgical treatment and prognosis evaluation.