This project focuses on early screening, early diagnosis and early treatment of peripheral lung cancer. Supported by 21 funds such as the Ministry of Science and Technology and the National Natural Science Foundation of China, this project lasted for 16 years. It targeted imaging examination of lung nodules, lung cancer screening, quantitative and qualitative diagnosis, imaging omics, artificial intelligence and molecular imaging and carried out a series of studies based on multimodal imaging on major clinical and basic scientific issues, and the correct diagnosis rate of early lung cancer reached 94.74%. The main innovative results are as follows: 1. Create a new method for detecting pulmonary nodules; optimize the detection and follow-up plan ① It is proposed for the first time to change the position target scan, and the diagnostic accuracy rate of lesions has increased from 44% to 90%.② Rib-suppression imaging technology and CT temporal subtraction technology were used for the first time to evaluate the detection and dynamic changes of pulmonary nodules.③ For the first time in the world, it has been proposed that the CT quantitative index of pure ground-glass density pulmonary nodules is related to the degree of lung tissue expansion.④ Optimize the follow-up scanning plan for GGN. 2. Propose a new method for classifying ground-glass nodules; formulate qualitative and quantitative standards for early lung cancer ① A new method for the definition and classification of GGN has been created for the first time in the world. It is proposed that the use of an intermediate window setting (-50WL/2WW) can achieve accurate classification and invasive judgment of GGN's real components (specificity 82.5%, accuracy 75.6%).② China took the lead in proposing important malignant signs of ground-glass nodules and established a predictive model for the qualitative diagnosis of lung cancer. It formed an important part of the update of international consensus and was introduced into the British Thoracic Society Guidelines and Canadian Lung Nodules Reporting Analysis System. 3. Lead imaging omics and artificial intelligence research to improve the clinical treatment level of ground-glass nodules. The earliest domestic development of GGN radiomics automatic segmentation technology, with recognition accuracy of 96.35%, and segmentation accuracy of 81.70%. For the first time, it has been proved that enhanced CT images do not affect the extraction of genomic characteristics. China is the first to establish a model to determine the infiltration of ground-glass lung adenocarcinoma, distant metastasis of NSCLC, and EGFR mutations. It is the first time in the world that an imaging model based on tumor and peritumoral tumors can accurately predict preoperative lymph node metastasis in early NSCLC (prediction accuracy AUC0.862). China took the lead in carrying out artificial intelligence research in medical imaging, and formulated the first international consensus on CT signs and labeling of lung nodules for artificial intelligence research. For the first time in the world, the ability of an artificial intelligence model for chest radiographs to detect ground-glass nodules has been explored. 4. Standardize the lung cancer screening path, build an early screening and early diagnosis system to carry out large-scale multi-center early lung cancer screening in Shanghai (14506 cases). It was found that the detection rate of lung nodules in Shanghai was 29.89%, and the true positive rate of lung nodules was 3.48%., the detection rate of lung cancer was 1.23%, and the proportion of stage I lung cancer was 81.09%. It took the lead in formulating an expert consensus on low-dose CT screening and clinical management of subsolid nodules, laying the foundation for large-scale lung cancer screening and the treatment of screened nodules in my country. 5. Focusing on precise diagnosis and treatment of lung cancer, building targeted molecular probes have built a series of multifunctional integrated diagnosis and treatment nanoprobes in terms of targeted identification of lung cancer tumor cells and diagnosis and treatment. For the first time, we have successfully constructed a dual targeting nanoprobe carrying magnetic resonance and radionuclide tracers; a multifunctional nanodiagnosis and treatment system loaded with doxorubicin and cisplatin using carbon nanohorns as a carrier; It was found that the morphology and active targeting of nanomaterials can affect the microscopic dynamic processes inside tumors, and then affect their uptake of nanomaterials, laying the foundation for efficient diagnosis and treatment of lung cancer. A total of more than 100 papers were published in this project, including 62 SCI papers, with a total IF of 237.43 points and 873 citations. A total of 17 doctorates and 39 masters have been trained. The research results of lung nodule examination technology and early diagnosis of lung cancer have been well promoted and applied in many major hospitals across the country. The results of the lung cancer screening program provide good suggestions for formulating guidelines for my country's lung cancer screening program.