Radiomics Nomogram for Early Diabetic Kidney Injury Prediction from Ultrasound Cohorts

This study develops and validates a radiomics-based clinical nomogram using routine renal ultrasound imaging and standard clinical data to predict early diabetic kidney injury in a community cohort. The nomogram, which integrates a radiomics signature with clinical determinants like glycated hemoglobin levels and diabetes duration, significantly outperforms clinical-only models in predicting early kidney injury. The study highlights the nomogram's clinical utility and potential as a non-invasive, cost-effective screening tool, offering a pragmatic approach for early intervention in community-based diabetic populations.