Synthetic CT Generation using a Diffusion Model: A Strategy for MRI-Based Attenuation Correction of PET Images
Abstract
Attenuation correction (AC) is critical for accurate positron emission tomography (PET). Hybrid systems with magnetic resonance imaging (MRI) used MRI-based AC (MRAC) methods. MRI does not provide photon attenuation information, making MRAC less accurate than computed tomography (CT)-based AC (CTAC). This thesis proposes diffusion model-based MRAC (DM-MRAC) for neurological PET/MRI. A diffusion model was trained to convert T1-weighted MRI to synthetic CTs for AC of neurological 18F-fluorodeoxyglucose-PET. Accuracy of DM-MRAC and Dixon-based MRAC was evaluated with respect to CTAC using relative percent difference (REL%) and absolute percent difference (ABS%). DM-MRAC outperformed Dixon-based MRAC with global REL% of -1.8 ± 2.2% (p=0.003) and ABS% of 1.9 ± 2.0% (p=0.001). Significant improvements (p < 0.0012) were observed in 10 brain regions in REL% and 5 in ABS%. This thesis shows DM-MRAC can reduce global and regional error in neurological PET/MRI, supporting DM-MRAC as a feasible improvement over Dixon-based MRAC.
Read the paper
Related studies
- Learning from First-Generation Qualitative Approaches in the IS Discipline: An Evolutionary View and Some Implications for Authors and Evaluators (PART 1/2)
- Commentary on the clinical and preclinical dosage limits of interstitially administered magnetic fluids for therapeutic hyperthermia based on current practice and efficacy models
- Prosthesis: New Technological Opportunities and Innovative Biomedical Devices
- The history and future of neural modeling for cochlear implants
- Ryanodine receptors: waking up from refractoriness
- Novel Transgenic Rabbit Model Sheds Light on the Puzzling Role of Matrix Metalloproteinase-12 in Atherosclerosis
Go deeper in Evidence Hub
Evidence Hub adds a study data sheet (design, sample size, evidence level, key numbers), an AI summary and a chat grounded in the paper's text.