2023 World Conference on Lung Cancer (Posters)-P1.27. Two- and Three-dimensional Radiomic Measurements for Clinical Stage IA Lung Adenocarcinoma: Less is More

P1.27. Two- and Three-dimensional Radiomic Measurements for Clinical Stage IA Lung Adenocarcinoma: Less is More - PDF(Slides)

Back to course

Pdf Summary

In this study, the authors aimed to compare the effectiveness of two-dimensional (2D) and three-dimensional (3D) radiomic measurements in predicting invasiveness in lung adenocarcinoma. They manually delineated pulmonary nodules along the boundary in every cross-sectional image to extract image features. A total of 818 patients were included in the primary cohort, which was divided into internal testing and validation cohorts. Additionally, 474 patients were examined within an independent validation cohort.<br /><br />The performance of consolidation tumor ratio (CTR) and radiomic models for 2D and 3D were assessed using receiver operating characteristic curves in these cohorts. The results showed that both CTR and radiomic models for 2D and 3D had comparable performance with areas under the curves (AUC). Furthermore, significant recurrence-free survival (RFS) and overall survival (OS) were observed in the primary cohort using cutoff values of 0.81 for 2D-CTR and 0.23 for 3D-CTR.<br /><br />Based on these results, the authors concluded that three-dimensional measurement does not provide any additional clinical benefit compared to two-dimensional measurement. This suggests that 2D radiomic measurements can be effectively used to assess prognosis and invasiveness in early lung adenocarcinoma. The study also included graphs depicting the segmentation and calculation of 2D-CTR, as well as the performance of CTR and radiomic models in different cohorts and the Kaplan-Meier survival analysis for RFS and OS.

Asset Subtitle

Meng-Min Wang

Meta Tag

Speaker Meng-Min Wang

Topic Early-Stage NSCLC: Innovation & New Technology

Keywords

two-dimensional radiomic measurements

three-dimensional radiomic measurements

invasiveness prediction

lung adenocarcinoma

pulmonary nodules

image features

consolidation tumor ratio

receiver operating characteristic curves

recurrence-free survival

overall survival