The central challenge is that MRI signal intensity reflects magnetic properties rather than the photon-attenuation behavior needed for PET reconstruction. Consequently, an MR image cannot simply be substituted for a CT attenuation map. MRI attenuation correction must translate MR-derived tissue information into attenuation coefficients, allowing PET reconstruction to account for how different tissues weaken emitted photons.
MR-based correction commonly relies on tissue classification. The attenuation map distinguishes categories such as soft tissue, bone, and air, then assigns each category an attenuation coefficient. Imaging hardware can also require representation in the map. These distinctions matter because treating all regions as equivalent would not capture the different ways tissues and hardware affect photon transmission during PET imaging.
Using a tissue-specific map supports more reliable interpretation of PET measurements alongside MRI findings. In cancer research, improved correction can strengthen lesion localization and standardized uptake value measurements, or SUVs, by reducing the mismatch between the physical imaging process and quantitative reconstruction. This is particularly important when assessing metabolic findings during treatment response studies.
A typical workflow begins with MR image acquisition, followed by either classification of tissues from those images or generation of a synthetic CT-like attenuation map. The resulting map represents soft tissue, bone, air, and relevant imaging hardware with assigned attenuation coefficients. PET reconstruction then uses this information to compensate for photon weakening, supporting combined anatomical and metabolic interpretation.
Tissue-classification methods derive attenuation information by assigning MR-identified regions to categories such as soft tissue, bone, or air. Synthetic CT-like approaches instead generate a map designed to resemble the attenuation information provided by CT. Both approaches supply coefficients for PET reconstruction, but they represent different strategies for converting MR information into attenuation-relevant data.
In oncology, corrected PET/MRI data can support tumor staging, treatment response assessment, and evaluation of radiotracers. The correction also enables quantitative studies that combine PET-derived metabolic information with MRI’s anatomical and functional information. This combination helps investigators relate tracer behavior to the location and detailed imaging characteristics of cancer-related findings.