Screening is an important tool for early detection of lung cancer. yang et al. reported that 76% of lung cancers detected by screening were early stage. Therefore planned screening of specific populations can detect early lung cancer. The number of new tumor cases in China amounts to millions every year, and if the diagnosis rate of early stage tumor can be increased by one percentage point, thousands of tumor patients can be treated timely and effectively, so it is significant to widely carry out screening and improve the rate of early tumor diagnosis. At present, the methods of screening are mainly as follows: 1. Sputum test: including sputum cytology examination, sputum immunolabeling, sputum PCR technology, etc. Sputum examination is convenient, non-invasive, easy to be accepted by patients, and suitable for tumors involving large bronchi, but its examination is affected by many factors, and its sensitivity and specificity are not high, especially for peripheral lung cancer, so the sensitivity of sputum screening needs to be further improved. 2.Serum tumor marker test: serum tumor markers are used to indirectly determine the existence of malignant lesions through the detection of specific substances secreted into the blood by the lesions. They include carcinoembryonic antigen (CEA), cytokeratin fragment antigen 21-1 (CYFRA21-1), squamous carcinoma antigen (SCCAG), tissue peptide antigen (TPA), neurospecific enolase (NSE), creatine phosphokinase-BB (CPK-BB) and so on. However, since the immunomarkers for different cell types of lung cancer often differ from each other, and even for the same cell type of lung cancer, it is difficult to find common immunomarkers; moreover, the current technology is being developed and perfected, and the sensitivity and specificity of the test itself are greatly restricted by technical conditions, and the data reported by various institutions also vary greatly, so there are no reliable markers that can be used for large-scale The general survey. 3.X-ray: It is the classical census method with the advantage of being economical, convenient and effective, and is still widely used. However, due to the limitation of resolution, lesions below 1 cm are often not detected, and even if they are detected, it is difficult to characterize them. In a control group of 5483 screening cases, it was found that CT found 8 times more lesions than X-ray, which means that 87% of the lesions found by CT could not be found on X-ray! So X-rays are of limited use. But with the widespread use of CR (computerized photography) and DR (computerized digital photography) and CAD (computer-aided detection and diagnosis system) being used as an adjunct to X-ray screening, their lesion detection rate is increasing. For developing countries like ours, the conditions are not yet available to carry out CT screening on a large scale, so using X-ray combined with CAD for screening is a more practical choice. 4.Low-dose spiral CT: It is the most sensitive and specific means of lung cancer screening! The dose of X-ray irradiation it receives is similar to that of ordinary X-rays, but its accuracy is very high, and it can detect tiny lesions of several millimeters in the lung, and the nature can be further clarified by more detailed target scans and other means, so it is also more acceptable. Some studies have pointed out that: the detection rate of non-calcified nodules is ten times higher in CT than in CR; lung cancer detected by CT is earlier staged and smaller in diameter than lung cancer detected by CR; the 5-year survival rate of lung cancer detected by CT is higher than that of lung cancer detected by CR; and the rate of surgically resectable lung cancer detected by CT is significantly higher than that detected by CR. Therefore, low-dose spiral CT should be the preferred and accurate method for lung cancer screening, and should be used as much as possible when economic conditions allow. The literature reports that the detection rate of X-ray is significantly lower than that of CT, and the number of lesions detected by low-dose spiral CT can be 8 times higher than that of ordinary X-ray. 76% of small subclinical lung cancers detected by low-dose CT cannot be shown on X-ray. Therefore, CT should be the screening tool of choice. The screening technique of CT is very important, and in 1999, the Japanese Lung Cancer Association published a guideline for CT screening, in which they considered it appropriate to use spiral CT with a layer thickness of 1-3 mm for routine lung cancer screening. In contrast, Shingo et al. showed that a layer thickness of 1 mm is more accurate for the identification of benign and malignant isolated lung nodules. Therefore, the current screening with a layer thickness of 10 mm will have a large number of misses. With the development of multi-row CT, the layer thickness is getting thinner and thinner, and now the layer thickness has been reduced to 1 mm for routine 16-row CT scans, allowing retrospective personalized reconstruction as needed. Regarding scanning conditions, different authors have reported different optimal doses ranging from 20 mA to 120 mA at constant voltage (120 Kv -140 Kv). Schoepf et al. suggested sweeping the entire chest at 120 kV at a dose of 10-40 mA (adjusted for individual body size). However, Zwirewich et al. compared images obtained at 40 mA with those obtained at 400 mA and found that low-dose CT showed lung parenchyma relatively satisfactorily in most cases, but failed to show subtle hairy glass signs (20%) and emphysema (11%) in some cases, and they suggested using 80-90 mA for the first HRCT scan and 40-50 mA for subsequent reviews. in the first HRCT scan and 40-50 mA in the subsequent review. Most of the nodules detected by CT screening are small, with subtle signs that do not necessarily show some of the typical signs of larger lung cancers, and further fine scans are needed for small nodules. The doctor proposed the “one target, three multi” small nodule examination plan. The “triple” refers to the use of multiple windows to show the lung interface and internal structure, multiple phase dynamic enhancement to evaluate the blood supply and density characteristics, and multiple post-processing techniques to show more signs and anatomical relationships. The diagnostic accuracy of this method is 95.7% compared to 68.6% for conventional CT, which is significantly better than conventional scanning. Imaging signs of lung cancer include morphology, density, and enhancement characteristics. Lobulation, burr, vacuolar sign, air bronchial sign, hairy glass sign, pleural depression sign, pulmonary vascular concentration sign, and obvious enhancement sign are all considered malignant signs, but early stage lung cancer has many specific manifestations that do not necessarily appear typical of larger lung cancers. In terms of morphology, there is a high rate of display of burr or burred margins, but a lower rate of occurrence of signs such as lobar signs and pleural depression signs. We know that the secondary lung lobular structures are about 1-3 cm, and the basis of lobular formation is firstly the obstruction of the framework structures in the lung, and uneven differentiation of the tumor in larger sizes is also involved in the formation of lobules. Therefore, if the tumor does not fill the whole lobule when it is small, it is not easy to form the lobar sign. As for the pleural depression sign, the occurrence of pleura being pulled is also low because the small nodules have not yet formed enough traction.