Search for a command to run...
Abstract Radiomics is reshaping quantitative imaging by converting biomedical images into high-dimensional biomarkers that capture biological, molecular, and functional processes. While its clinical adoption—particularly in oncology—has accelerated, preclinical applications are rapidly expanding, driven by high-resolution imaging technologies and the need for robust quantitative frameworks to support translational research. Preclinical imaging spans micro-computed tomography (micro-CT), high-field magnetic resonance imaging (MRI), micro-positron emission tomography (micro-PET), micro-single-photon emission computed tomography (SPECT), and emerging modalities such as optical and photoacoustic imaging, applied across diverse experimental models from murine systems and patient-derived xenografts (PDX) to zebrafish and 3D cellular platforms. Despite its promise, preclinical radiomics faces challenges including variability in acquisition and reconstruction, limited sample sizes, lack of standardized workflows, and difficulties in biological interpretation. This review provides a systematic overview of current literature, organized by imaging modality and biological model, highlighting methodological patterns, shared limitations, and emerging opportunities. Special attention is given to harmonization strategies, publicly available datasets, and integrated software platforms that enable reproducible pipelines and lower technical barriers. Ultimately, this work delineates key directions to enhance reproducibility and accelerate the translation of preclinical radiomics into clinically meaningful applications.