Generation of a Four-Class Attenuation Map for MRI-Based Attenuation Correction of PET Data in the Head Area Using a Novel Combination of STE/Dixon-MRI and FCM Clustering.

Purpose: The aim of this study is to generate a four-class magnetic resonance imaging (MRI)-based attenuation map (μ-map) for attenuation correction of positron emission tomography (PET) data of the head area using a novel combination of short echo time (STE)/Dixon-MRI and a dedicated image segmentation method.

Procedures: MR images of the head area were acquired using STE and two-point Dixon sequences. μ-maps were derived from MRI images based on a fuzzy C-means (FCM) clustering method along with morphologic operations. Quantitative assessment was performed to evaluate generated MRI-based μ-maps compared to X-ray computed tomography (CT)-based μ-maps.

Results: The voxel-by-voxel comparison of MR-based and CT-based segmentation results yielded an average of more than 95 % for accuracy and specificity in the cortical bone, soft tissue, and air region. MRI-based μ-maps show a high correlation with those derived from CT scans (R (2) > 0.95).

Conclusions: Results indicate that STE/Dixon-MRI data in combination with FCM-based segmentation yields precise MR-based μ-maps for PET attenuation correction in hybrid PET/MRI systems.