Deep Learning Model Enables Contrast-Free Renal Volumetry on Non-Contrast MRI

A deep learning model trained on paired contrast-enhanced and non-contrast MRI segmented kidneys on non-contrast T1 with Dice 0.950. Total kidney volume had mean absolute error 8.76 mL (2.5% of mean) and concordance correlation 0.983, enabling contrast-free renal volumetry.

• Cortical volume fraction was overestimated (74.7% vs. 72.1%) and medullary fraction underestimated (21.5% vs. 24.3%), while total kidney volumetry was highly accurate.
• This retrospective single-center study validated model performance only in kidneys without major structural distortion.

Automated summary

RadPigeon summaries are original and for information only. They are not clinical advice.