Radiomic analysis of medical imaging for classification in oncology and recommendations for clinically initiated research: a literature review.

Background and Objective: Radiomics is a thriving field that aims to enhance clinical decision-making by enabling the noninvasive, quantitative characterization of lesions on medical images. Despite thousands of studies being published in this field, the adaptation of radiomics in routine clinical practice remains challenging due to the complexity of analytical steps and issues in reproducibility. The goal of this review was to facilitate translation by bridging the gap between radiomics research and clinical applications in oncology. Methods: A comprehensive literature search was conducted of major databases for literature on the application of radiomics to clinical classification in oncology published from January 1, 2005, to November 30, 2025. We reviewed eligible articles, summarized and discussed their content, and provided recommendations for each step of the radiomic analysis workflow. Key Content and Findings: The literature review generated 13 key recommendations for improving the quality, reliability, and reproducibility of radiomic models for disease characterization in clinical oncology. These recommendations covered important aspects, including data quality assurance, robust feature selection techniques, open data, and model sharing. Conclusions: By considering these recommendations, researchers and clinicians may improve the clinical applicability of radiomics, aiding its gradual incorporation into routine practice in oncology. The integration of radiomics into clinical settings holds the potential to enhance patient care and contribute to the advancement of personalized medicine.