Lung cancer is second causes of new cancer case in worldwide with differences in clinicopathological parameters.CT imaging plays a critical role in early diagnosis, stage prediction, and follow-up of the patients having lung cancer The main aim of this study was to discovery of histopathology and overall staging in lung cancer by using intra-tumoral heterogeneity quantification in CT scan.

402 patient from the cancer image archives with different histopathology (Large cell 110, squamous cell carcinoma 140, adenocarcinoma 48, not otherwise specified 62 and not specified 42) and stage (I 92, II 37, IIIa 104 and IIIb 168) were included in this study. All lesion were segmented by using automatic region growing algorithm. Image were preprocessed by Laplacian of Gaussian (LOG) filter with different sigma value (0.5 to 5 with step 0.5 to produce fine, medium and coarse textures). Fowling image preprocessing 790 features including...

The proposed framework provided an area under the curve (AUC) value of 61.19 and 63.28 for classification in histopathology and overall staging respectively, with 10-fold cross validation.

The results in this study indicate the feasibility of intra-tumoral heterogeneity quantification for discovery of histopathology and overall staging in CT scan of lung cancer patient. The proposed method can effectively predict histopathology and overall stagingin lung cancer by image biomarker and advanced machine learning methods. Quantification of intra-tumoral heterogeneity information provide valuable clinical information which can be used as additional information for improving diagnostic and prognostic model in oncology.

ADDIN EN.REFLIST 1. Shiri I, Rahmim A, Ghaffarian P, Geramifar P, Abdollahi H, Bitarafan-Rajabi A. The impact of image reconstruction settings on 18F-FDG PET radiomic features: multi-scanner phantom and patient studies. European Radiology. 2017:1-12. 2. Junior JRF, Cipriano FEG, Fabro AT, Koenigkam-Santos M, de Azevedo-Marques PM, editors. Radiomics-Based Recognition of Metastatic and Histopathological Patterns of Lung Cancer. European Congress on Computational Methods in Applied Sciences and Engineering; 2017: Springer. 3. Aerts HJ, Velazquez ER, Leijenaar RT, Parmar C, Grossmann P, Cavalho S, et al. Decoding...