Fig. 1: Quantitative accuracy evaluated with global physical metrics on external scanners and external tracers. | Nature Communications

Fig. 1: Quantitative accuracy evaluated with global physical metrics on external scanners and external tracers.

From: Using domain knowledge for robust and generalizable deep learning-based CT-free PET attenuation and scatter correction

Fig. 1

Quantitative accuracy of the DL ASC-PET images generated with our proposed method, to the CT ASC-PET on the cross-scanner (a–c) and cross-tracer (d–f) settings, evaluated with global physical metrics including normalized root mean squared error (NRMSE), peak signal-to-noise ratio (PSNR) and structural similarity index measurement (SSIM). Data are presented as mean values +/− SD. Sample size: a–c: Siemens Vision 450 (n = 51), Siemens Vision 600 (n = 62), UI uMI 780 (n = 98), GE Discovery MI (n = 104); d–f: 68Ga-FAPI (n = 7), 68Ga-DOTA-TATE (n = 17), 68Ga-DOTA-TOC (n = 8), 18F-PSMA (n = 12).

Back to article page