探讨高分辨率CT下肺磨玻璃结节(GGN)影像学特征与肺腺癌病理新分类的相关性，评估高分辨率CT下GGN对术后病理类型的预测价值。

分析2017年1月至2019年12月于我院胸外科接受胸腔镜肺叶、肺段、肺楔形切除83例患者临床资料，均随访6个月，按病理结果分为原位腺癌(AIS)和不典型瘤样增生(AAH)、微浸润腺癌(MIA)、浸润腺癌(IAC)三组，对不同分类的患者结节数量、病灶位置、形态、病变内部、血管特征及周边征象、平均大小、平均CT值、最大CT值进行分析。采用二元Logistic回归分析定量资料和定性资料，并保存预测概率值，得出各组间敏感度、特异度及临界值。

三组间平均大小、平均CT值、最大CT值、体积、质量差异有统计学意义(P<0.05)，且IAC组平均大小、体积和质量明显大于AAH、AIS组和MIA组，平均CT值、最大CT值明显小于AAH、AIS和MIA组(P<0.05)。三组间瘤-肺界面、病灶形态比较差异无统计学意义(P>0.05)，而边缘、病变内部征象、血管及周边征象比较差异有统计学意义(P<0.05)，且IAC组GGN的边缘、病变内部征象、血管及周边征象均大于其他两组(P<0.05)。AAH、AIS与MIA分组的危险因素有平均大小、平均CT值及毛刺征，其MIA所得曲线下面积、敏感度、特异度、临界值分别为(0.890、0.925、0.775、0.280)；MIA与IAC分组的危险因素有平均大小、平均CT值、最大CT值、胸膜凹陷征，其IAC所得曲线下面积、敏感度、特异度、临界值分别为(0.940、0.899、0.965、0.690)。所有患者术后均随访6个月，11例非腺癌患者均无1例复发，而72例肺腺癌患者有9例(12.50%)复发，1例(1.39%)死亡。

通过GGN的CT特征及定量资料分析，预测GGN肺腺癌的病理分类，其预测价值高于单因素。

To investigate the correlation between the imaging features of lung ground glass nodules (GGN) and the new classification of lung adenocarcinoma under high-resolution CT, and to evaluate the predictive value of GGN under high-resolution CT on postoperative pathological types.

Retrospectively analyzed the clinical data of 83 patients who underwent thoracoscopic lobectomy, lung segment, and wedge wedge resection at the Department of Thoracic Surgery, Our Hospital from January 2017 to December 2019, and were followed up for 6 months. Cancer (AIS) and atypical neoplasia (AAH), microinfiltrating adenocarcinoma (MIA), infiltrating adenocarcinoma (IAC), the number of nodules, lesion location, morphology, internal lesions, and vascular characteristics of patients with different classifications And the surrounding signs, average size, average CT value, maximum CT value were analyzed. Binary logistic regression was used to analyze quantitative data and qualitative data, and the predicted probability values were saved to obtain the sensitivity, specificity and critical value between groups.

The differences in the average size, average CT value, maximum CT value, volume, and mass between the three groups were statistically significant (P<0.05), and the average size, volume, and mass of the IAC group were significantly larger than those of the AAH, AIS, and MIA groups, and the average CT value The maximum CT value was significantly smaller than the AAH, AIS and MIA groups (P<0.05). There was no significant difference in tumor lung interface and focus morphology among the three groups (P>0.05), but there was significant difference in edge, internal signs of lesions, vascular and peripheral signs (P<0.05), and the edge, internal signs of lesions, vascular and peripheral signs of GGN in IAC group were larger than those in the other two groups (P<0.05). The risk factors for AAH, AIS, and MIA grouping are average size, average CT value, and glitch sign. The area under the curve, sensitivity, specificity, and critical value obtained by MIA are (0.890, 0.925, 0.775, 0.280); MIA and IAC. The risk factors for grouping included the average size, average CT value, maximum CT value, and pleural depression. The area under the curve, sensitivity, specificity, and critical value obtained by IAC were (0.940, 0.899, 0.965, 0.690). All 83 patients were followed up for 6 months after operation. None of the 11 non-adenocarcinoma patients relapsed, while 9 (12.50%) of 72 lung adenocarcinoma patients relapsed and 1 (1.39%) died.

The CT features and quantitative data analysis of GGN were used to predict the pathological classification of GGN lung adenocarcinoma.