Deep-learning denoising may salvage low-dose 18F-FDG breast PET/MRI image quality

A deep-learning denoising algorithm applied to simulated low-dose (90% reduced) 18F-FDG breast PET/MRI significantly improved image quality, with lesions judged equivalent or slightly inferior to full-dose images in 89.9% of reads. Quantitative noise decreased (p<0.0001) and tum…

• Three readers assessed 23 women with primary invasive breast cancer; PET data were undersampled to simulate a 37 MBq injected activity.
• Readers preferred denoised images over non-denoised low-dose images in 76.8% of reads for overall preference.
• This was a pilot secondary analysis; prospective validation of the low-dose protocol has not been performed.

Automated summary

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