Radiomic-Based Quantitative CT Analysis of Pure Ground-Glass Nodules to Predict the Invasiveness of Lung Adenocarcinoma

Abstract

Objectives: To investigate the performance of radiomic-based quantitative analysis on CT images in predicting invasiveness of lung adenocarcinoma manifesting as pure ground-glass nodules (pGGNs). Methods: A total of 275 lung adenocarcinoma cases, with 322 pGGNs resected surgically and confirmed pathologically, from January 2015 to October 2017 were enrolled in this retrospective study. All nodules were split into training and test cohorts randomly with a ratio of 4:1 to establish models to predict between pGGN-like adenocarcinoma in situ (AIS)/minimally invasive adenocarcinoma (MIA) and invasive adenocarcinoma (IVA). Radiomic feature extraction was performed using Pyradiomics with semi-automatically segmented tumor regions on CT scans that were contoured with an in-house plugin for 3D-Slicer. Random forest (RF) and support vector machine (SVM) were used for feature selection and predictive model building in the training cohort. Three different predictive models containing conventional, radiomic, and combined models were built on the basis of the selected clinical, radiological, and radiomic features. The predictive performance of each model was evaluated through the receiver operating characteristic curve (ROC) and the area under the curve (AUC). The predictive performance of two radiologists (A and B) and our radiomic predictive model were further investigated in the test cohort to see if radiomic predictive model could improve radiologists' performance in prediction between pGGN-like AIS/MIA and IVA. Results: Among 322 nodules, 48 (14.9%) were AIS and 102 (31.7%) were MIA with 172 (53.4%) for IVA. Age, diameter, density, and nine meaningful radiomic features were selected for model building in the training cohort. Three predictive models showed good performance in prediction between pGGN-like AIS/MIA and IVA (AUC > 0.8, P < 0.05) in both training and test cohorts. The AUC values in the test cohort were 0.824 (95% CI, 0.723–0.924), 0.833 (95% CI, 0.733–0.934), and 0.848 (95% CI, 0.750–0.946) for conventional, radiomic, and combined models, respectively. The predictive accuracy was 73.44 and 59.38% for radiologist A and radiologist B in the test cohort and was improved dramatically to 79.69 and 75.00% with the aid of our radiomic predictive model. Conclusion: The predictive models built in our study showed good predictive power with good accuracy and sensitivity, which provided a non-invasive, convenient, economic, and repeatable way for the prediction between IVA and AIS/MIA representing as pGGNs. The radiomic predictive model outperformed two radiologists in predicting pGGN-like AIS/MIA and IVA, and could significantly improve the predictive performance of the two radiologists, especially radiologist B with less experience in medical imaging diagnosis. The selected radiomic features in our research did not provide more useful information to improve the combined predictive mode...

Exploration of an Integrative Prognostic Model of Radiogenomics Features With Underlying Gene Expression Patterns in Clear Cell Renal Cell Carcinoma

“…In our study, the predictive efficacy of the elected prognostic related radiomics features based on training set were found to be in accordance with some of the reference research above ( 30 , 33 , 34 , 36 ). However, a lot of former studies have concentrated on the performance of textural features of radiographic images, which may lack a comprehensive explanation of the biological mechanism and potential biomolecular features of the disease.…”

Radiomic signature based on CT imaging to distinguish invasive adenocarcinoma from minimally invasive adenocarcinoma in pure ground-glass nodules with pleural contact

“…Zhang et al [ 33 ] demonstrated that histogram parameters, combined with an evaluation of morphological characteristics, exhibited good diagnostic performances in discriminating AIS/MIA from IAC, appearing as pGGNs, The AUC, sensitivity and specificity of the predictive model was 0.896, 0.794, and 0.914, respectively. Similarly, for the prediction between AIS/MIA and IAC representing as pGGNs, Xu et al [ 34 ] showed the predictive radiomics models built in study (AUC 0.833;95% CI, 0.733–0.934) which provided a good predictive power. Besides, Sun et al [ 35 ] developed a radiomics-based Rad-score utilized as a biomarker for the invasiveness-predicted evaluation in patients with pGGNs (AUC 0.72; 95% CI, 0.63–0.81).…”

CT-based radiomics for differentiating invasive adenocarcinomas from indolent lung adenocarcinomas appearing as ground-glass nodules: A systematic review

“…The number of patients/nodules ranged between 34/34 and 794/886. Eight studies [15,[17][18][19][20][21][22][23] included only mixed ground-glass nodules (MGGNs), seven [24][25][26][27][28][29][30] only pure ground-glass nodules (PGGNs), and thirteen [14,16,[31][32][33][34][35][36][37][38][39][40][41] both of MGGNs and PGGNs.…”

“…The ROIs were segmented manually in 20 studies [14,15,18,21,22,[24][25][26][27]29,[31][32][33][34][36][37][38][39][40][41], semi-automatically in six studies [17,19,20,28,30,35] and automatically in two studies [16,23] (Table 2). Multiple segmentations, more than one person segmented the ROIs, were conducted in 24 studies.…”

“…Multiple segmentations, more than one person segmented the ROIs, were conducted in 24 studies. Eleven studies [14,15,17,19,24,28,29,31,35,37,39] reported that the persons who segmented the ROIs were blinded to the histopathological results of the tumors. ROIs in the study by Wu et al [15] included not only the gross tumor volume, but also the solid regions of the GGN and the peri-nodular regions.…”