Image quality in radiomics for enhanced medical diagnostics: A metrology overview with a critical approach

Radiomics is a crucial discipline for personalized medicine, specifically by enhancing diagnostic accuracy and treatment decisions. However, the reliability of radiomics features heavily relies on the quality of medical imaging as well as on the standardization methods for the acquisition and the processing protocols. Significant challenges include the variability introduced by different equipment, operator handling, and patient-related factors. All above factors may affect the consistency and the clinical applicability of radiomics outcomes. This work highlights the critical aspects of image quality and standardized image processing techniques as pivotal to improve the reproducibility of radiomics data. Furthermore, the need for rigorous metrology approaches in image evaluation is emphasized to ensure that radiomics may fulfil its potential in clinical settings. To boost the research in this field, an overview of the recent literature is presented, and a critical analysis is performed, by considering all stages of the radiomics workflow. The final goal is to highlight the influence quantities potentially affecting the reliability of the radiomics features extraction.