Because of its high aggressiveness and lethality, early detection and accurate characterization of lung cancer are among the most investigated challenges in the last years. Biomedical imaging techniques are an enabling tool for lung cancer assessment that strongly impacts on the decision-making process in daily clinical practice. Their role is a developing process that aims to provide predictive imaging biomarkers and a valid solution for personalized medicine. Within this context, radiomic approach, a relatively new solution that consists of a features-based characterization of the tumor, is showing promising results. It requires the mining of vast arrays of quantitative features derived from digital images and has opened to encouraging perspectives. Regardless of the interest in such a solution, a suboptimal standardization and lack of definitive results emerge. LuCIFEx presents the design and development of an automated pipeline for a non-invasive in-vivo identification and characterization of Non Small Cell Lung Cancer (NSCLC). It is devised to be a support for radiologists and physicians in the treatment decision phase and to speed up the diagnostic process. The developed pipeline exploits input data from routinely acquired biomedical images. From these images, we obtained a segmentation of the tumor lesion allowing the accurate textural features computation inside the Volume Of Interest (VOI), thus providing information for the characterization of lung lesion through Machine Learning algorithms.