Related Experiment Video
Updated: Jun 13, 2025
Author Spotlight: Advancing 3D Modeling for Enhanced Diagnosis and Treatment of Pulmonary Nodules in Early-Stage Lung Cancer
Published on: October 13, 2023
Study on the value of 3D visualization in differentiating IA and non-IA pulmonary ground-glass nodules
1Ganzhou Institute of Medical Imaging, Ganzhou Key Laboratory of Medical Imaging and Artificial Intelligence, Medical Imaging Center, Ganzhou People's Hospital, The Affiliated Ganzhou Hospital of Nanchang University, Ganzhou Hospital-Nanfang Hospital, Southern Medical University, 16th Meiguan Avenue, Ganzhou 341000, PR China; Clinical Research Center for Medical Imaging in Jiangxi Province, Nanchang 330000, PR China.
Objective:
To determine the most effective diagnostic markers and their associated thresholds for Ground-glass nodules (GGN) for identification of invasive adenocarcinoma (IA) and non-IA (including atypical adenomatous hyperplasia (AAH), adenocarcinoma in situ (AIS), and minimally invasive adenocarcinoma (MIA)), and to explore the application in preoperative surgical evaluation.
Methods And Methods:
A total of 126 cases, confirmed by pathology, were retrospectively analyzed. 70 cases were classified as the IA group, while the non-IA group consisted of cases of AAH, AIS, and MIA, with a total of 56 cases. The study compared the differences in demographic, morphological, and three-dimensional (3D) quantitative parameters between the two groups.
Results:
There were statistically significant differences in various signs such as air bronchogram, lobulation, pleural indentation, spiculation, shape, and margin between the two groups. Additionally, Statistical significance was observed in all 3D quantitative parameters for both groups. Notably, when 3D volume of lesions exceeded 447 mm3, the sensitivity to predict IA was 81% with specificity at 69%. Utilizing multivariate logistic regression analysis, it was identified that the independent predictive value in discriminating between IA and non-IA lies with both the 3D volume and solid proportion. Combining these two indexes significantly improved the prediction accuracy (AUC = 0.826).
Conclusions:
Reasonable utilization of 3D visualization technology can effectively aid in distinguishing between IA and non-IA. When coupled with clinical data and CT signs, this technique holds vital importance in directing the evaluation of surgical interventions prior to surgery.
