biostudies-literatureWu JUniversity of Texas MD Anderson Cancer Center Lung Moon Shot ProgramCancer Research UKNCI NIH HHSU.S. Department of Health & Human Services | National Institutes of HealthEngineering and Physical Sciences Research Council787-798https://www.ebi.ac.uk/biostudies/studies/S-EPMC8612063biostudies-literatureUnknown3Radiomics refers to the high-throughput extraction of quantitative features from radiological scans and is widely used to search for imaging biomarkers for prediction of clinical outcomes. Current radiomic signatures suffer from limited reproducibility and generalizability, because most features are dependent on imaging modality and tumor histology, making them sensitive to variations in scan protocol. Here, we propose novel radiological features that are specially designed to ensure compatibility across diverse tissues and imaging contrast. These features provide systematic characterization of tumor morphology and spatial heterogeneity. In an international multi-institution study of 1,682 patients, we discover and validate four unifying imaging subtypes across three malignancies and two mNature machine intelligenceRadiological tumor classification across imaging modality and histology.PMC8612063K99 CA21866719732R00 CA218667R01 CA193730EP/N014588/1R01 CA222512R01 CA233578Wu JNeal JWLi CDiehn MShirato HGensheimer MPadda SLi RWei YJaffray DHeymach JLoo BWSchonlieb CBWakelee HPrice SJKato FfalseRadiological tumor classification across imaging modality and histology.Radiomics refers to the high-throughput extraction of quantitative features from radiological scans and is widely used to search for imaging biomarkers for prediction of clinical outcomes. Current radiomic signatures suffer from limited reproducibility and generalizability, because most features are dependent on imaging modality and tumor histology, making them sensitive to variations in scan protocol. Here, we propose novel radiological features that are specially designed to ensure compatibility across diverse tissues and imaging contrast. These features provide systematic characterization of tumor morphology and spatial heterogeneity. In an international multi-institution study of 1,682 patients, we discover and validate four unifying imaging subtypes across three malignancies and two m2021-01-01T00:00:00Z2021 Sep2026-05-31T08:09:27.077Z2025-04-05T15:22:02.863ZS-EPMC86120633484119510.1038/s42256-021-00377-0