A Multi-Classification Model for Predicting the Invasiveness of Lung Adenocarcinoma Presenting as Pure Ground-Glass Nodules

Abstract

ObjectivesTo establish a multi-classification model for precisely predicting the invasiveness (pre-invasive adenocarcinoma, PIA; minimally invasive adenocarcinoma, MIA; invasive adenocarcinoma, IAC) of lung adenocarcinoma manifesting as pure ground-glass nodules (pGGNs).MethodsBy the inclusion and exclusion criteria, this retrospective study enrolled 346 patients (female, 297, and male, 49; age, 55.79 ± 10.53 (24-83)) presenting as pGGNs from 1292 consecutive patients with pathologically confirmed lung adenocarcinoma. A total of 27 clinical were collected and 1409 radiomics features were extracted by PyRadiomics package on python. After feature selection with L2,1-norm minimization, logistic regression (LR), extra w(ET) and gradient boosting decision tree (GBDT) were used to construct the three-classification model. Then, an ensemble model of the three algorithms based on model ensemble strategy was established to further improve the classification performance.ResultsAfter feature selection, a hybrid of 166 features consisting of 1 clinical (short-axis diameter, ranked 27th) and 165 radiomics (4 shape, 71 intensity and 90 texture) features were selected. The three most important features are wavelet-HLL_firstorder_Minimum, wavelet-HLL_ngtdm_Busyness and square_firstorder_Kurtosis. The hybrid-ensemble model based on hybrid clinical-radiomics features and the ensemble strategy showed more accurate predictive performance than other models (hybrid-LR, hybrid-ET, hybrid-GBDT, clinical-ensemble and radiomics-ensemble). On the training set and test set, the model can obtain the accuracy values of 0.918 ± 0.022 and 0.841, and its F1-scores respectively were 0.917 ± 0.024 and 0.824.ConclusionThe multi-classification of invasive pGGNs can be precisely predicted by our proposed hybrid-ensemble model to assist patients in the early diagnosis of lung adenocarcinoma and prognosis.

Effective Invasiveness Recognition of Imbalanced Data by Semi-Automated Segmentations of Lung Nodules

“…Sun et al [33] provided multivariate logistic regression analysis for PGGNs with LASSO. Song et al [34] compared the results of using logistic regression (LR), extra trees (ET), and a gradient boosting decision tree (GBDT) with selected radiomics features.…”

“…Hence, a potential question for effectiveness differences of the proposed semi-automated segmentation and fully automated ones needs to be replied. For this question, three recent fully automated segmentation methods, including Mask-RCNN [7], Unet [24], and SeResUnet [34] were compared with the proposed method SSTM. Figure 20 reveals that the proposed SSTM achieves much better dice than the fully automated methods, reaching a dice improvement of 392.3%.…”

The identification of hub genes associated with pure ground glass nodules using weighted gene co-expression network analysis

“…Known as ground glass nodules (GGNs), these nodules mainly exists in bronchial and vascular margins in the lung [ 2 ]. GGN can be further divided into part-solid GGN and pGGN, according to whether solid components are existed in the lesion [ 3 ]. If an invasive pGGN persist for a long term, an early malignant tumor may be associated with this condition, so it is crucial to distinguish the invasiveness of pGGNs [ 4 ].…”

The Identification of Hub Genes Associated with Pure Ground Glass Nodules Using Weighted Gene Co-Expression Network Analysis

“…With the widely application of computed tomography with low dose,a gradually increasing number of pGGNs have been detected [1] .Known as ground glass nodules (GGNs), these nodules mainly exists in bronchial and vascular margins in the lung [2] .GGN can be further divided into part-solid GGN and pGGN,according to whether solid components are existed in the lesion [3] . If an invasive pGGN persist for a long term, an early malignant tumor may be associated with this condition, so it is crucial to distinguish the invasiveness of pGGNs [4] .…”