Super-resolution deep learning boosts brain FLAIR quality while preserving WMH volumes

Super-resolution DL reconstruction improved brain FLAIR quality (noise, SNR, sharpness; p<0.001) in 36 volunteers, preserving white matter hyperintensity volumes matching high-res standard, without distorting lesion measures.

• Prospective study in 36 healthy volunteers; image quality was quantitatively and qualitatively superior with SR-DLR vs. Gaussian and denoising DLR (all p<0.001).
• Automated WMH volumes did not differ significantly across reconstructions and closely matched a dedicated high-resolution acquisition.
• Limitation: small volunteer cohort without extensive pathology; no external validation of the reconstruction method reported.

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