Dataset Information

A triple-classification for the evaluation of lung nodules manifesting as pure ground-glass sign: a CT-based radiomic analysis.

ABSTRACT:

Objectives

To construct a noninvasive radiomics model for evaluating the pathological degree and an individualized treatment strategy for patients with the manifestation of ground glass nodules (GGNs) on CT images.

Methods

The retrospective primary cohort investigation included patients with GGNs on CT images who underwent resection between June 2015 and June 2020. The intratumoral regions of interest were segmented semiautomatically, and radiomics features were extracted from the intratumoral and peritumoral regions. After feature selection by ANOVA, Max-Relevance and Min-Redundancy (mRMR) and Least Absolute Shrinkage and Selection Operator (Lasso) regression, a random forest (RF) model was generated. Receiver operating characteristic (ROC) analysis was calculated to evaluate each classification. Shapley additive explanations (SHAP) was applied to interpret the radiomics features.

Results

In this study, 241 patients including atypical adenomatous hyperplasia (AAH) or adenocarcinoma in situ (AIS) (n = 72), minimally invasive adenocarcinoma (MIA) (n = 83) and invasive adenocarcinoma (IAC) (n = 86) were selected for radiomics analysis. Three intratumoral radiomics features and one peritumoral feature were finally identified by the triple RF classifier with an average area under the curve (AUC) of 0.960 (0.963 for AAH/AIS, 0.940 for MIA, 0.978 for IAC) in the training set and 0.944 (0.955 for AAH/AIS, 0.952 for MIA, 0.926 for IAC) in the testing set for evaluation of the GGNs.

Conclusion

The triple classification based on intra- and peritumoral radiomics features derived from the noncontrast CT images had satisfactory performance and may be used as a noninvasive tool for preoperative evaluation of the pure ground-glass nodules and developing of individualized treatment strategies.

SUBMITTER: Yu Z

PROVIDER: S-EPMC9327229 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:ObjectiveTo develop a radiomic model based on low-dose CT (LDCT) to distinguish invasive adenocarcinomas (IAs) from adenocarcinoma in situ/minimally invasive adenocarcinomas (AIS/MIAs) manifesting as pure ground-glass nodules (pGGNs) and compare its performance with conventional quantitative and semantic features of LDCT, radiomic model of standard-dose CT, and intraoperative frozen section (FS).MethodsA total of 147 consecutive pathologically confirmed pGGNs were divided into primary cohort (43 IAs and 60 AIS/MIAs) and validation cohort (19 IAs and 25 AIS/MIAs). Logistic regression models were built using conventional quantitative and semantic features, selected radiomic features of LDCT and standard-dose CT, and intraoperative FS diagnosis, respectively. The diagnostic performance was assessed by area under curve (AUC) of receiver operating characteristic curve, sensitivity, and specificity.ResultsThe AUCs of quantitative-semantic model, radiomic model of LDCT, radiomic model of standard-dose CT, and FS model were 0.879 (95% CI, 0.801-0.935), 0.929 (95% CI, 0.862-0.971), 0.941 (95% CI, 0.876-0.978), and 0.884 (95% CI, 0.805-0.938) in the primary cohort and 0.897 (95% CI, 0.768-0.968), 0.933 (95% CI, 0.815-0.986), 0.901 (95% CI, 0.773-0.970), and 0.828 (95% CI, 0.685-0.925) in the validation cohort. No significant difference of the AUCs was found among these models in both the primary and validation cohorts (all p > 0.05).ConclusionThe LDCT-based quantitative-semantic score and radiomic signature, with good predictive performance, can be pre-operative and non-invasive biomarkers for assessing the invasive risk of pGGNs in lung cancer screening.Advances in knowledgeThe LDCT-based quantitative-semantic score and radiomic signature, with the equivalent performance to the radiomic model of standard-dose CT, can be pre-operative predictors for assessing the invasiveness of pGGNs in lung cancer screening and reducing excess examination and treatment.

Project description:ObjectivesTo investigate the value of radiomics based on CT imaging in predicting invasive adenocarcinoma manifesting as pure ground-glass nodules (pGGNs).MethodsThis study enrolled 395 pGGNs with histopathology-confirmed benign nodules or adenocarcinoma. A total of 396 radiomic features were extracted from each labeled nodule. A Rad-score was constructed with the least absolute shrinkage and selection operator (LASSO) in the training set. Multivariate logistic regression analysis was conducted to establish the radiographic model and the combined radiographic-radiomics model. The predictive performance was validated by receiver operating characteristic (ROC) curve. Based on the multivariate logistic regression analysis, an individual prediction nomogram was developed and the clinical utility was assessed.ResultsFive radiomic features and four radiographic features were selected for predicting the invasive lesions. The combined radiographic-radiomics model (AUC 0.77; 95% CI, 0.69-0.86) performed better than the radiographic model (AUC 0.71; 95% CI, 0.62-0.81) and Rad-score (AUC 0.72; 95% CI, 0.63-0.81) in the validation set. The clinical utility of the individualized prediction nomogram developed using the Rad-score, margin, spiculation, and size was confirmed in the validation set. The decision curve analysis (DCA) indicated that using a model with Rad-score to predict the invasive lesion would be more beneficial than that without Rad-score and the clinical model.ConclusionsThe proposed radiomics-based nomogram that incorporated the Rad-score, margin, spiculation, and size may be utilized as a noninvasive biomarker for the assessment of invasive prediction in patients with pGGNs.Key points• CT-based radiomics analysis helps invasive prediction manifested as pGGNs. • The combined radiographic-radiomics model may be utilized as a noninvasive biomarker for predicting invasive lesion for pGGNs. • Radiomics-based individual nomogram may serve as a vital decision support tool to identify invasive pGGNs, obviating further workup and blind follow-up.

Project description: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 model's performance.

Project description:BackgroundPure ground-glass nodules (pGGNs) with pleural contact (P-pGGNs) comprise not only invasive adenocarcinoma (IAC), but also minimally invasive adenocarcinoma (MIA). Radiomics recognizes complex patterns in imaging data by extracting high-throughput features of intra-tumor heterogeneity in a non-invasive manner. In this study, we sought to develop and validate a radiomics signature to identify IAC and MIA presented as P-pGGNs.MethodsIn total, 100 patients with P-pGGNs (69 training samples and 31 testing samples) were retrospectively enrolled from December 2012 to May 2018. Imaging and clinical findings were also analyzed. In total, 106 radiomics features were extracted from the 3D region of interest (ROI) using computed tomography (CT) imaging. Univariate analyses were used to identify independent risk factors for IAC. The least absolute shrinkage and selection operator (LASSO) method with 10-fold cross-validation was used to generate predictive features to build a radiomics signature. Receiver-operator characteristic (ROC) curves and calibration curves were used to evaluate the predictive accuracy of the radiomics signature. Decision curve analyses (DCA) were also conducted to evaluate whether the radiomics signature was sufficiently robust for clinical practice.ResultsUnivariate analysis showed significant differences between MIA (N = 47) and IAC (N = 53) groups in terms of patient age, lobulation signs, spiculate margins, tumor size, CT values and relative CT values (all P < 0.05). ROC curve analysis showed, when MIA was identified from IAC, that the critical value of tumor length diameter (TLD) was1.39 cm and the area under the ROC curve (AUC) was 0.724 (sensitivity = 0.792, specificity = 0.553). The critical CT value on the largest axial plane (CT-LAP) was - 597.45 HU, and the AUC was 0.666 (sensitivity = 0.698, specificity= 0.638). The radiomics signature consisted of seven features and exhibited a good discriminative performance between IAC and MIA, with an AUC of 0.892 (sensitivity = 0.811, specificity 0.719), and 0.862 (sensitivity = 0.625, specificity = 0.800) in training and testing samples, respectively.ConclusionsOur radiomics signature exhibited good discriminative performance in differentiating IAC from MIA in P-pGGNs, and may offer a crucial reference point for follow-up and selective surgical management.

Project description:Objective: To evaluate whether radiomic features extracted from intra and peri-nodular lesions can enhance the ability to differentiate between invasive adenocarcinoma (IA), minimally invasive adenocarcinoma (MIA), and adenocarcinoma in situ (AIS) manifesting as ground-glass nodule (GGN). Materials and Methods: This retrospective study enrolled 120 patients with a total of 121 pathologically confirmed lung adenocarcinomas (85 IA and 36 AIS/MIA) from January 2015 to May 2019. The recruited patients were randomly divided into training (84 nodules) and validation sets (37 nodules), with a ratio of 7:3. The minority group in the training set was balanced by the synthetic minority over-sampling (SMOTE) method. The intra-, peri-nodular, and gross region of interests (ROI) were delineated with manual annotation. Image features were quantitatively extracted from each ROI on CT images. The minimum redundancy maximum relevance (mRMR) feature ranking method and the least absolute shrinkage and selection operator (LASSO) classifier were used to eliminate unnecessary features. The intra- and peri-nodular radiomic features were combined to produce the gross radiomic signature. A combined clinical-radiomic model was constructed by multivariable logistic regression analysis. The predicted performances of different models were evaluated using receiver operating curve (ROC) and calibration curve. Results: The gross radiomic signature (AUC: training set = 0.896; validation set = 0.876) showed a good ability to discriminate the invasiveness of adenocarcinoma, comparing to intra-nodular (AUC: training set = 0.862; validation set = 0.852) or peri-nodular radiomic signature (AUC: training set = 0.825; validation set = 0.820). The AUC of the combined clinical-radiomic model was 0.917 for the training and 0.876 for the validation cohort, respectively. Conclusions: The gross radiomic signature of intra- and peri-nodular regions improved the prediction ability and aided predicting the invasiveness of lung adenocarcinoma appearing as GGN.

Project description:BackgroundPatients with consistent lung pure ground-glass nodules (pGGNs) have a high incidence of lung adenocarcinoma that can be classified as adenocarcinoma in situ (AIS), minimally invasive adenocarcinoma (MIA), or invasive adenocarcinoma (IAC). Regular follow-up is recommended for AIS and MIA, while surgical resection should be considered for IAC. This study sought to develop a multi-parameter prediction model to increase the diagnostic accuracy in discriminating between IAC and AIS or MIA.MethodsThe training data set comprised consecutive patients with lung pGGNs who underwent resection from January to December 2017 at the Zhongshan Hospital. Of the 370 resected pGGNs, 344 were pathologically confirmed to be AIS, MIA, or IAC and were included in the study. The 26 benign pGGNs were excluded. We compared differences in the clinical features (e.g., age and gender), the content of serum tumor biomarkers, the computed tomography (CT) parameters (e.g., nodule size and the maximal CT value), and the morphologic characteristics of nodules (e.g., lobulation, spiculation, pleura indentation, vacuole sign, and normal vessel penetration or abnormal vessel) between the pathological subtypes of AIS, MIA, and IAC. An abnormal vessel was defined as "vessel curve" or "vessel enlargement". Statistical analyses were performed using the chi-square test, analysis of variance (ANOVA), and rank test. The IAC prediction model was constructed via a multivariate logistical regression. Our prediction model for lung pGGNs was further validated in a data set comprising consecutive patients from multiple medical centers in China from July to December 2018. In total, 345 resected pGGNs were pathologically diagnosed as lung adenocarcinoma in the validation data set.ResultsIn the training data set, patients with pGGNs ≥10 mm in size had a high incidence (74.5%) of IAC. The maximal CT value of IAC [-416.1±121.2 Hounsfield unit (HU)] was much higher than that of MIA (-507.7±138.0 HU) and AIS (-602.6±93.3 HU) (P<0.001). IAC was more common in pGGNs that displayed any of the following CT manifestations: lobulation, spiculation, pleura indentation, vacuole sign, and vessel abnormality. The IAC prediction model was constructed using the parameters that were assessed as risk factors (i.e., the nodule size, maximal CT value, and CT signs). The receiver operating characteristic (ROC) analysis showed that the area under the curve (AUC) of this model for diagnosing IAC was 0.910, which was higher than that of the AUC for nodule size alone (0.891) or the AUC for the maximal CT value alone (0.807) (P<0.05, respectively). A multicenter validation data set was used to validate the performance of our prediction model in diagnosing IAC, and our model was found to have an AUC of 0.883, which was higher than that of the AUC of 0.827 for the module size alone model or the AUC of 0.791 for the maximal CT value alone model (P<0.05, respectively).ConclusionsOur multi-parameter prediction model was more accurate at diagnosing IAC than models that used only nodule size or the maximal CT value alone. Thus, it is an efficient tool for identifying the IAC of malignant pGGNs and deciding if surgery is needed.