Dataset Information

Computed tomography-based radiomics machine learning models for prediction of histological invasiveness with sub-centimeter subsolid pulmonary nodules: a retrospective study.

ABSTRACT: To improve the accuracy of preoperative diagnoses and avoid over- or undertreatment, we aimed to develop and compare computed tomography-based radiomics machine learning models for the prediction of histological invasiveness using sub-centimeter subsolid pulmonary nodules. Three predictive models based on radiomics were built using three machine learning classifiers to discriminate the invasiveness of the sub-centimeter subsolid pulmonary nodules. A total of 203 sub-centimeter nodules from 177 patients were collected and assigned randomly to the training set (n = 143) or test set (n = 60). The areas under the curve of the predictive models were 0.743 (95% confidence interval CI [0.661-0.824]) for the logistic regression, 0.828 (95% CI [0.76-0.896]) for the support vector machine, and 0.917 (95% CI [0.869-0.965]) for the XGBoost classifier models in the training set, and 0.803 (95% CI [0.694-0.913]), 0.726 (95% CI [0.598-0.854]), and 0.874 (95% CI [0.776-0.972]) in the test set, respectively. In addition, the decision curve showed that the XGBoost model added more net benefit within the range of 0.06 to 0.93.

SUBMITTER: Zhang H

PROVIDER: S-EPMC9838201 | biostudies-literature | 2023

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Computed tomography-based radiomics machine learning models for prediction of histological invasiveness with sub-centimeter subsolid pulmonary nodules: a retrospective study.

Zhang Haochuan H Wang Shixiong S Deng Zhenkai Z Li Yangli Y Yang Yingying Y Huang He H

PeerJ 20230110

To improve the accuracy of preoperative diagnoses and avoid over- or undertreatment, we aimed to develop and compare computed tomography-based radiomics machine learning models for the prediction of histological invasiveness using sub-centimeter subsolid pulmonary nodules. Three predictive models based on radiomics were built using three machine learning classifiers to discriminate the invasiveness of the sub-centimeter subsolid pulmonary nodules. A total of 203 sub-centimeter nodules from 177 p ...[more]

PMID: 36643621

Similar Datasets

Project description:ObjectivesTo evaluate the clinical and non-contrast computed tomography (CT) features of patients with benign pulmonary subsolid nodules (SSNs) with a solid component ≤ 5 mm and their development trends via follow-up CT.MethodsWe retrospectively collected 436 data from patients who had SSNs with a solid component ≤ 5 mm, including 69 with absorbable benign SSNs (AB-SSNs), 70 with nonabsorbable benign SSNs (NB-SSNs), and 297 with malignant SSNs (M-SSNs). Models 1, 2, and 3 for distinguishing the different types of SSNs were then developed and validated.ResultsPatients with AB-SSNs were younger and exhibited respiratory symptoms more frequently than those with M-SSNs. The frequency of nodules detected during follow-up CT was in the following order: AB-SSNs > NB-SSNs > M-SSNs. NB-SSNs were smaller than M-SSNs, and ill-defined margins were more frequent in AB-SSNs than in NB-SSNs and M-SSNs. Benign SSNs exhibited irregular shape, target sign, and lower CT values more frequently compared to M-SSNs, whereas the latter demonstrated bubble lucency more commonly compared to the former. Furthermore, AB-SSNs showed more thickened interlobular septa and satellite lesions than M-SSNs and M-SSNs had more pleural retraction than AB-SSNs (all p < 0.017). The three models had AUCs ranging 0.748-0.920 and 0.790-0.912 in the training and external validation cohorts, respectively. A follow-up CT showed nodule progression in four benign SSNs.ConclusionsThe three SSN types have different clinical and imaging characteristics, with some benign SSNs progressing to resemble malignancy.Critical relevance statementA good understanding of the imaging features and development trends of benign SSNs may help reduce unnecessary follow-up or interventions. This retrospective study explores the CT characteristics of benign SSNs with a solid component ≤ 5 mm by comparing AB-SSNs, NB-SSNs, and M-SSNs and delineates their development trends via follow-up CT.Key points1. Different subsolid nodule types exhibit distinct clinical and imaging features. 2. A miniscule number of benign subsolid nodules can progress to resemble malignancy. 3. Knowing the clinical and imaging features and development trends of benign subsolid nodules can improve management.

Project description:BackgroundAccurate evaluation of pulmonary nodule malignancy is important for lung cancer management. This current study aimed to develop risk models for small solid and subsolid pulmonary nodules based on clinical and quantitative radiomics features.MethodsThis study enrolled 5-20 mm pulmonary nodules detected on thoracic high-resolution computed tomography (HRCT), which were all confirmed pathologically. There were 548 solid nodules (242 malignant vs. 306 benign) and 623 subsolid nodules (SSNs 519 malignant vs. 104 benign). Relevant clinical characteristics were recorded. The CT image prior to the initial treatment was chosen for manual segmentation of the targeted nodule using the ITK-SNAP software. Subsequently, the marked image was processed to quantitatively extract 1218 radiomics features using PyRadiomics. We performed five-fold cross-validation to select potential predictors from clinical and radiomics features using the LASSO method and to evaluate the performance of the established models. In total, four types of models were tried: random forest, XGBOOST, SVM, and logistic models. The established models were compared with the Mayo model.ResultsLung cancer risk models were developed among four nodule groups: all nodules (410 benign vs. 761 malignant; 1:1.86), nodules ≤10 mm (185 benign vs. 224 malignant; 1:1.21), solid nodules (306 benign vs. 242 malignant; 1.26:1), and SSNs (104 benign vs. 104 malignant; 1:1 matched). Significant clinical and radiomics predictors were selected for each group. The accuracy, area under the ROC curve, sensitivity, and specificity of the best model on validation dataset was 0.86, 0.91, 0.93, 0.73 for all nodules (XGBOOST), 0.82, 0.90, 0.86, 0.76 for nodules ≤10 mm (XGBOOST), 0.80, 0.89, 0.78, 0.82 for solid nodules (XGBOOST) and 0.70, 0.73, 0.73, 0.67 for SSNs (Random Forest). Except for the SSN models, the established clinical-radiomics models were superior to the Mayo model.ConclusionsPredictive models based on both clinical and radiomics features can be used to assess the malignancy of small solid and subsolid pulmonary nodules, even for nodules that are 10 mm or smaller.

Dataset Information

Computed tomography-based radiomics machine learning models for prediction of histological invasiveness with sub-centimeter subsolid pulmonary nodules: a retrospective study.

Publications

Computed tomography-based radiomics machine learning models for prediction of histological invasiveness with sub-centimeter subsolid pulmonary nodules: a retrospective study.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets