ObjectiveTo evaluate the application value of three-dimensional (3D) reconstruction in preoperative surgical diagnosis of new classification criteria for lung adenocarcinoma, which is helpful to develop a deep learning model of artificial intelligence in the auxiliary diagnosis and treatment of lung cancer.MethodsThe clinical data of 173 patients with ground-glass lung nodules with a diameter of ≤2 cm, who were admitted from October 2018 to June 2020 in our hospital were retrospectively analyzed. Among them, 55 were males and 118 were females with a median age of 61 (28-82) years. Pulmonary nodules in different parts of the same patient were treated as independent events, and a total of 181 subjects were included. According to the new classification criteria of pathological types, they were divided into pre-invasive lesions (atypical adenomatous hyperplasia and and adenocarcinoma in situ), minimally invasive adenocarcinoma and invasive adenocarcinoma. The relationship between 3D reconstruction parameters and different pathological subtypes of lung adenocarcinoma, and their diagnostic values were analyzed by multiplanar reconstruction and volume reconstruction techniques.ResultsIn different pathological types of lung adenocarcinoma, the diameter of lung nodules (P<0.001), average CT value (P<0.001), consolidation/tumor ratio (CTR, P<0.001), type of nodules (P<0.001), nodular morphology (P<0.001), pleural indenlation sign (P<0.001), air bronchogram sign (P=0.010), vascular access inside the nodule (P=0.005), TNM staging (P<0.001) were significantly different, while nodule growth sites were not (P=0.054). At the same time, it was also found that with the increased invasiveness of different pathological subtypes of lung adenocarcinoma, the proportion of dominant signs of each group gradually increased. Meanwhile, nodule diameter and the average CT value or CTR were independent risk factors for malignant degree of lung adenocarcinoma.ConclusionImaging signs of lung adenocarcinoma in 3D reconstruction, including nodule diameter, the average CT value, CTR, shape, type, vascular access conditions, air bronchogram sign, pleural indenlation sign, play an important role in the diagnosis of lung adenocarcinoma subtype and can provide guidance for personalized therapy to patients in clinics.
Lung adenocarcinoma is a prevalent histological subtype of non-small cell lung cancer with different morphologic and molecular features that are critical for prognosis and treatment planning. In recent years, with the development of artificial intelligence technology, its application in the study of pathological subtypes and gene expression of lung adenocarcinoma has gained widespread attention. This paper reviews the research progress of machine learning and deep learning in pathological subtypes classification and gene expression analysis of lung adenocarcinoma, and some problems and challenges at the present stage are summarized and the future directions of artificial intelligence in lung adenocarcinoma research are foreseen.