(HealthDay News) — Photon-counting detector (PCD) computed tomography (CT) with deep learning noise reduction may improve spatial resolution for viewing findings in multiple myeloma compared with energy-integrating detector (EID) CT, according to a study published online Sept. 6 in Radiology.
Francis I. Baffour, M.D., from Mayo Clinic in Rochester, Minnesota, and colleagues prospectively enrolled and scanned 27 adult participants who underwent a whole-body EID CT scan with a PCD CT system in ultra-high-resolution mode at matched radiation dose at an academic medical center between April and July 2021. EID CT images were reconstructed with Br44 and Br64 kernels at 2-mm section thickness; PCD CT images were reconstructed with both Br44 and Br64 kernels at 2-mm section thickness and Br44 and Br76 kernels at 0.6-mm section thickness. A convolutional neural network was used to denoise the thinner PCD CT images. To detect findings reflecting multiple myeloma, two radiologists scored PCD CT images relative to EID CT.
The researchers found that in blinded assessment of 2-mm images, PCD CT demonstrated improvement versus EID CT in viewing lytic lesions, intramedullary lesions, fatty metamorphosis, and pathologic fractures. For viewing all four pathologic abnormalities, improvement was also demonstrated for the 0.6-mm PCD CT images with convolutional neural network denoising; in 21 of 27 participants, one or more lytic lesions were detected compared with the 2-mm EID CT images.
“A clinical photon-counting detector CT in ultra-high-resolution mode with and without deep learning noise reduction algorithm demonstrated superior performance in showing multiple myeloma lesions relative to energy-integrating detector CT,” the authors write.
One author disclosed financial ties to the pharmaceutical and medical device industries, including Siemens Healthineers, which owns the evaluated system and partially funded the study.