Objectives: The preoperative differentiation of adenocarcinomas in situ, minimally invasive adenocarcinoma, and invasive adenocarcinoma using computed tomography (CT) is crucial for guiding clinical management decisions. However, accurately classifying ground-glass nodules poses a significant challenge. Incorporating quantitative intratumoral heterogeneity scores may improve the accuracy of this ternary classification. Materials and methods: In this multicenter retrospective study, we developed ternary classification models by leveraging insights from both base and stacking ensemble machine learning models, incorporating intratumoral heterogeneity scores along with clinical-radiological features to distinguish adenocarcinomas in situ, minimally invasive adenocarcinoma, and invasive adenocarcinoma. The machine learning models were trained, and final model selection depended on maximizing the macro-average area under the curve (macro-AUC) in both the internal and external validation sets. Results: Data from 802 patients from three centers were divided into a training set (n = 477) and an internal test set (n = 205), in a 7:3 ratio, with an additional external validation set comprising 120 patients. The stacking classifier exhibited superior performance relative to the other models, achieving macro-AUC values of 0.7850 and 0.7717 for the internal and external validation sets, respectively. Moreover, an interpretability analysis utilizing the Shapley Additive Explanation identified four key features of this ternary classification: intratumoral heterogeneity score, nodule size, nodule type, and age. Conclusion: The stacking classifier, recognized as the optimal algorithm for integrating the intratumoral heterogeneity score and clinical-radiological features, effectively served as a ternary classification model for assessing the invasiveness of lung adenocarcinoma in chest CT images. Critical relevance statement: Our stacking classifier integrated intratumoral heterogeneity scores and clinical-radiological features to improve the ternary classification of lung adenocarcinoma invasiveness (adenocarcinomas in situ/minimally invasive adenocarcinoma/invasive adenocarcinoma), aiding in precise diagnosis and clinical decision-making for ground-glass nodules. Key Points: The intratumoral heterogeneity score effectively assessed the invasiveness of lung adenocarcinoma. The stacking classifier outperformed other methods for this ternary classification task. Intratumoral heterogeneity score, nodule size, nodule type, and age predict invasiveness. Graphical

The novel grading system developed by the International Association for the Study of Lung Cancer (IASLC) for clinical stage IA lung adenocarcinomas has demonstrated remarkable prognostic capabilities. Notably, tumors classified as grade 3 have been associated with poor prognostic outcomes, thereby playing a crucial role in the formulation of personalized surgical strategies. The objective of this study is to develop a radiomics nomogram that utilizes the optimal volume of interest (VOI) derived from high-resolution CT (HRCT) scans to accurately predict the presence of grade 3 tumors in patients with clinical IA lung adenocarcinomas. In this multi-center, large-population study, clinical, pathological, and HRCT imaging data from 1418 patients who were pathologically diagnosed with lung adenocarcinomas were retrospectively collected. The data was obtained from four hospital databases between January 2018 and May 2022. From this patient cohort, 1206 individuals were screened from three databases and randomly divided into training and internal validation datasets in a 7:3 ratio. An additional dataset consisting of 212 individuals was used for external validation dataset. Radiomics features were extracted from HRCT images at various scales, including VOI^−2mm, VOI ^entire, VOI ^+2mm, and VOI ^+4mm. To reduce dimensionality, select relevant features, and build radiomics signatures, the maximal redundancy minimal relevance (mRMR) and least absolute shrinkage and selection operator (LASSO) algorithm were utilized. Univariate and multivariate logistic regression analyses were conducted to identify independent clinic-radiological (Clin-Rad) predictors. Receiver operating characteristic (ROC) curves and corresponding area under the curve (AUC) were used to evaluate the diagnostic efficiency. A nomogram predicting the risk of grade 3 in clinical stage IA lung adenocarcinoma was constructed based on multivariate logistic regression, combining independent predictors and the optimal radiomics signatures. Multivariate logistic regression revealed that males exhibited a higher prevalence of grade 3 tumors, and solid nodules were frequently observed through radiological assessments. The utilization of radiomics features extracted from the VOI ^entire resulted in significant improvements in predictive performance, as evidenced by AUC values of 0.900 (0.880-0.942), 0.885 (0.824-0.946), and 0.888 (0.782-0.993) for the training, internal validation, and external validation datasets, respectively. Furthermore, the nomogram that combined VOI ^entire -based radiomics signatures and Clin-Rad characteristics, exhibited remarkable predictive performance. This was indicated by AUC values of 0.910(0.873-0.942), 0.891 (0.845-0.937), and 0.905 (0.846-0.964) for the training, internal validation, and external validation datasets, respectively. The extraction of radiomics features from both the indented and peri-tumoral regions does not offer any additional benefits in predicting grade 3 tumors according to the IASLC system. However, when combining the VOI ^entire-based radiomics model with Clin-Rad characteristics, the resulting integrated nomogram exhibited remarkable predictive performance.

Introduction: This study evaluated the efficacy of radiomic analysis with optimal volumes of interest (VOIs) on computed tomography images to preoperatively differentiate invasive mucinous adenocarcinoma (IMA) from non-mucinous adenocarcinoma (non-IMA) in patients with incidental pulmonary nodules (IPNs). Methods: This multicenter, large-scale retrospective study included 1383 patients with IPNs, 110 (8%) of whom were pathologically diagnosed with IMA postoperatively. Radiomic features were extracted from multi-scale VOI subgroups (VOI ^{−2 mm} , VOI ^entire , VOI ^{+ 2 mm} , and VOI ^{+ 4 mm} ). Resampling methods, specifically, the synthetic minority oversampling technique, addressed the imbalance between the majority (IMA) and minority (non-IMA) groups. Radiomic features were identified using the least absolute shrinkage and selection operator algorithm. Radscores were calculated by linearly combining the selected features with their weights. A combined nomogram integrating the optimal VOI-based radiomic model with the image-finding classifier was constructed. Results: Bubble lucency and lower lobe predominance were significant in establishing an image-finding classifier to differentiate between IMA and non-IMA in IPNs, achieving an area under the curve (AUC) value of 0.684 (0.568-0.801). Across all radiomic models, IMA had a higher Radscore than did non-IMA. Specifically, the VOI + 2 mm-based radiomic model exhibited the highest performance, with an AUC of 0.832 (0.753-0.911). The combined nomogram outperformed the recognized image-finding classifier and radiomic models, achieving an AUC of 0.850 (0.776-0.925). Conclusion: A nomogram that combines a recognized image-finding classifier with an optimal VOI-based radiomic model effectively predicts IMA in IPNs, aiding physicians in developing comprehensive treatment strategies.