Multiclassifier fusion based on radiomics features for the prediction of benign and malignant primary pulmonary solid nodules

Background To test the ability of a multiclassifier model based on radiomics features to predict benign and malignant primary pulmonary solid nodules. Methods Computed tomography (CT) images of 342 patients with primary pulmonary solid nodules confirmed by histopathology or follow-up were retrospectively analyzed. The region of interest (ROI) of the images was delineated, and the radiomics features of the lesions were extracted. The feature weight was calculated using the relief feature selection algorithm. Based on the selected features, five classifier models were constructed: support vector machine (SVM), random forest (RF), logistic regression (LR), extreme learning machine (ELM), and K-nearest neighbor (KNN). The precision, recall rate, and area under the receiver operating characteristic curve (AUC) were used to evaluate the prediction performance of each classifier. The prediction result of each classifier was first weighted, and then all the prediction results were fused to predict the nodule type of unknown images. The prediction precision, recall rate, and AUC of the fusion classifier and single classifier were compared. Cross-validation was used to evaluate the generalization of the fusion classifier, and t- and F-tests were performed on the five classifiers and fusion classifier. Results For each ROI, 450 features in four major categories were extracted and were analyzed using the relief feature selection algorithm. According to the weights, 25 highly repetitive and nonredundant stable features that played a major role in pulmonary nodule classification were selected. The fusion classifier's prediction performance (prediction precision =92.0%, AUC =0.915) was superior to those of SVM (prediction precision =75.3%, AUC =0.740), RF (prediction precision =89.1%, AUC =0.855), LR (prediction precision =68.4%, AUC =0.681), ELM (prediction precision =87.0%, AUC =0.830), and KNN (prediction precision =77.1%, AUC =0.702). The fusion classifier showed the best null hypothesis performance in the t-test (P=0.035) and F-test (P=0.036). Conclusions The multiclassifier fusion model based on radiomics features had high prediction value for benign and malignant primary pulmonary solid nodules.

[1]  Sumit K. Shah,et al.  Pulmonary nodule characterization: a comparison of conventional with quantitative and visual semi-quantitative analyses using contrast enhancement maps. , 2006, European journal of radiology.

[2]  Andre Dekker,et al.  Radiomics: the process and the challenges. , 2012, Magnetic resonance imaging.

[3]  Patrick Granton,et al.  Radiomics: extracting more information from medical images using advanced feature analysis. , 2012, European journal of cancer.

[4]  V. Goh,et al.  Assessment of primary colorectal cancer heterogeneity by using whole-tumor texture analysis: contrast-enhanced CT texture as a biomarker of 5-year survival. , 2013, Radiology.

[5]  Yasuji Oshiro,et al.  Kurtosis and skewness assessments of solid lung nodule density histograms: differentiating malignant from benign nodules on CT , 2013, Japanese Journal of Radiology.

[6]  V. Goh,et al.  Non-small cell lung cancer: histopathologic correlates for texture parameters at CT. , 2013, Radiology.

[7]  R. Thornhill,et al.  Quantitative CT texture and shape analysis: Can it differentiate benign and malignant mediastinal lymph nodes in patients with primary lung cancer? , 2015, European Radiology.

[8]  Marcelo Costa Oliveira,et al.  Using 3D Texture and Margin Sharpness Features on Classification of Small Pulmonary Nodules , 2016, 2016 29th SIBGRAPI Conference on Graphics, Patterns and Images (SIBGRAPI).

[9]  Huiling Lu,et al.  Pulmonary Nodule Detection Model Based on SVM and CT Image Feature-Level Fusion with Rough Sets , 2016, BioMed research international.

[10]  Balaji Ganeshan,et al.  CT texture analysis can help differentiate between malignant and benign lymph nodes in the mediastinum in patients suspected for lung cancer , 2016, Acta radiologica.

[11]  Paul Kinahan,et al.  Radiomics: Images Are More than Pictures, They Are Data , 2015, Radiology.

[12]  Jian Shen,et al.  Medical image classification based on multi-scale non-negative sparse coding , 2017, Artif. Intell. Medicine.

[13]  D. Dong,et al.  Subtype discrimination of lung adenocarcinoma manifesting as ground glass nodule based on radiomics , 2017 .

[14]  Sameer Antani,et al.  Local-global classifier fusion for screening chest radiographs , 2017, Medical Imaging.

[15]  Di Dong,et al.  2D and 3D CT Radiomics Features Prognostic Performance Comparison in Non-Small Cell Lung Cancer , 2017, Translational oncology.

[16]  Lianfang Tian,et al.  Automatic benign and malignant classification of pulmonary nodules in thoracic computed tomography based on RF algorithm , 2018, IET Image Process..

[17]  X. Qi,et al.  A radiomics nomogram for preoperatively predicting prognosis of patients in hepatocellular carcinoma , 2018 .

[18]  Shu-Ju Tu,et al.  Localized thin-section CT with radiomics feature extraction and machine learning to classify early-detected pulmonary nodules from lung cancer screening , 2018, Physics in medicine and biology.

[19]  Jingyun Shi,et al.  Imaging phenotype using radiomics to predict dry pleural dissemination in non-small cell lung cancer. , 2019, Annals of translational medicine.

[20]  Imaging phenotype using radiomics to predict dry pleural dissemination in non-small cell lung cancer , 2019 .

[21]  Jiazhou Wang,et al.  Application of radiomics signature captured from pretreatment thoracic CT to predict brain metastases in stage III/IV ALK-positive non-small cell lung cancer patients. , 2019, Journal of thoracic disease.

[22]  Mohamed Elhoseny,et al.  Optimal feature level fusion based ANFIS classifier for brain MRI image classification , 2018, Concurr. Comput. Pract. Exp..