Deep Learning Denoising Improves and Homogenizes Patient [F]FDG PET Image Quality in Digital PET/CT.

Given the constant pressure to increase patient throughput while respecting radiation protection, global body PET image quality (IQ) is not satisfactory in all patients. We first studied the association between IQ and other variables, in particular body habitus, on a digital PET/CT. Second, to improve and homogenize IQ, we evaluated a deep learning PET denoising solution (Subtle PET) using convolutional neural networks. We analysed retrospectively in 113 patients visual IQ (by a 5-point Likert score in two readers) and semi-quantitative IQ (by the coefficient of variation in the liver, CV) as well as lesion detection and quantification in native and denoised PET. In native PET, visual and semi-quantitative IQ were lower in patients with larger body habitus ( < 0.0001 for both) and in men vs. women ( ≤ 0.03 for CV). After PET denoising, visual IQ scores increased and became more homogeneous between patients (4.8 ± 0.3 in denoised vs. 3.6 ± 0.6 in native PET; < 0.0001). CV were lower in denoised PET than in native PET, 6.9 ± 0.9% vs. 12.2 ± 1.6%; < 0.0001. The slope calculated by linear regression of CV according to weight was significantly lower in denoised than in native PET ( = 0.0002), demonstrating more uniform CV. Lesion concordance rate between both PET series was 369/371 (99.5%), with two lesions exclusively detected in native PET. SUV and SUV of up to the five most intense native PET lesions per patient were lower in denoised PET ( < 0.001), with an average relative bias of -7.7% and -2.8%, respectively. DL-based PET denoising by Subtle PET allowed [F]FDG PET global image quality to be improved and homogenized, while maintaining satisfactory lesion detection and quantification. DL-based denoising may render body habitus adaptive PET protocols unnecessary, and pave the way for the improvement and homogenization of PET modalities.