Metrics and radiomics

In recent years, medicine has witnessed an increase in the digitization of information generated during the clinical routine and the constant development of algorithms for analyzing biomedical images. From this point of view, the interest of the medical and scientific community in the world of Radiomics has grown, a “new science in oncology” which integrates radiological data with clinical data coming from the various players on the multidisciplinary cancer board. The concept of Radiomics is based on the assumption that biomedical images contain information which cannot be perceived by the human eye but which can be extracted by means of advanced mathematical analyses. By processing a huge amount of data, Radiomics aim to develop a diagnostic and predictive model for a tailor-made medical care, with the objective to find the best possible treatment for each patient.

It is evident that in this context the extraction of metrics represents a crucial element, which requires the use of solid, consolidated and well-documented tools.

Within the iTA Mining platform proprietary algorithms were implemented for the extraction of radiomic metrics (according to the indications of the Image Biomarker Standardization Initiative – IBSI – working group), developed in collaboration with DataMind.
Furthermore, given the complexity of the topic and the platform’s customization feature, it was decided to integrate predefined plugins that call scripts created in Python, R and Matlab and include extractors from external libraries already widely present on the market and used by the scientific community (Pyradiomics, Moddicom, LIFEx, …).