Using CT texture analysis to differentiate between peripheral lung cancer and pulmonary inflammatory pseudotumor

Abstract

Background: This study is to distinguish peripheral lung cancer and pulmonary inflammatory pseudotumor using CT-radiomics features extracted from PET/CT images. Methods: In this study, the standard 18F-fluorodeoxyglucose positron emission tomography/ computed tomography (18 F-FDG PET/CT) images of 21 patients with pulmonary inflammatory pseudotumor (PIPT) and 21 patients with peripheral lung cancer were retrospectively collected. The dataset was used to extract CT-radiomics features from regions of interest (ROI), The intra-class correlation coefficient (ICC) was used to screen the robust feature from all the radiomic features. Using, then, statistical methods to screen CT-radiomics features, which could distinguish peripheral lung cancer and PIPT. And the ability of radiomics features distinguished peripheral lung cancer and PIPT was estimated by receiver operating characteristic (ROC) curve and compared by the Delong test. Results: A total of 435 radiomics features were extracted, of which 361 features showed relatively good repeatability (ICC ≥ 0.6). 20 features showed the ability to distinguish peripheral lung cancer from PIPT. these features were seen in 14 of 330 Gray-Level Co-occurrence Matrix features, 1 of 49 Intensity Histogram features, 5 of 18 Shape features. The area under the curves (AUC) of these features were 0.731 ± 0.075, 0.717, 0.748 ± 0.038, respectively. The P values of statistical differences among ROC were 0.0499 (F9, F20), 0.0472 (F10, F11) and 0.0145 (F11, Mean4). The discrimination ability of forming new features (Parent Features) after averaging the features extracted at different angles and distances was moderate compared to the previous features (Child features). Conclusion: Radiomics features extracted from non-contrast CT based on PET/CT images can help distinguish peripheral lung cancer and PIPT.

Analysis of the value of enhanced CT combined with texture analysis in the differential diagnosis of pulmonary sclerosing pneumocytoma and atypical peripheral lung cancer: a feasibility study

“…Different from subjective image feature analysis, CT texture analysis can extract a variety of feature parameters from the image texture that cannot be distinguished by the naked eye and quantitatively analyse the heterogeneity of lesion tissue structure, and tumour heterogeneity is one of the important features that distinguishes malignant tumours from normal tissues or benign lesions [ 18 ]. Studies have shown that texture analysis based on CT images has good application prospects for the diagnosis of benign and malignant pulmonary lesions, pathological classification and staging of lung cancer [ 19 – 21 ]. The ERT is an integrated machine learning algorithm, and its base estimator is a decision tree.…”

CT Texture Analysis of Pulmonary Neuroendocrine Tumors—Associations with Tumor Grading and Proliferation

“…Concerning contrast media application, texture features might slightly differ between native CT scans and contrast-enhanced CT scans. A recent study employed texture analysis in native CT scans derived from PET-CT to discriminate peripheral lung cancer and inflammatory pseudotumor with promising results [ 36 ]. Moreover, reconstruction algorithms of the CT images seem to influence radiomics features [ 37 ].…”

Pre-Processing Method for Contouring the Uptake Levels of [18F] FDG for Enhanced Specificity of PET Imaging of Solitary Hypermetabolic Pulmonary Nodules

“…This approach of relying on extraction of features from a primary image is typical of radiomics, where image transformation techniques are used for advanced analysis, which can potentially reveal additional disease features and facilitate classification of neoplastic lesions. Therefore, in the context of radiomics, the PCUL-FDG method can be treated as this type of analysis, using the characteristics of shapes to assess the character of pulmonary nodules [ 14 , 15 , 16 , 17 ].…”