PT  - JOURNAL ARTICLE
AU  - Kaylynn Pinder
AU  - Michelle Gruchot
AU  - Scott Leonard
AU  - Marci Messina
AU  - Gary Dillehay
TI  - Continued evaluation of a complete µ-map generation in PET/MR Breast imaging.
DP  - 2021 May 01
TA  - Journal of Nuclear Medicine
PG  - 173--173
VI  - 62
IP  - supplement 1
4099  - http://jnm.snmjournals.org/content/62/supplement_1/173.short
4100  - http://jnm.snmjournals.org/content/62/supplement_1/173.full
SO  - J Nucl Med2021 May 01; 62
AB  - 173Objectives: Attenuation correction (AC) is necessary for data quantification in PET imaging. PET reconstructions without AC appear to show areas of low attenuation (patient surface and lungs) as having increased uptake. Traditional methods of generating a transmission-based AC map (µ-map) are unavailable with PET/MR imaging. Instead, MR imaging sequences (DIXON) are utilized to generate the required µ-map. Previous experiments demonstrated the creation of an incomplete µ-map if only the breast coils were utilized (per clinical diagnostic breast MR protocol). A complete µ-map was generated when an additional body matrix coil was placed on the subject’s back. But, the additional coil limits the size of the subject that can be imaged due MR bore clearance issues. Siemens mMR scientists proposed and modified the DIXON based MRAC sequence (permit multiple averages) to allow for creation of complete µ-maps utilizing only the internal body coil. When first testing this solution in 2019, a complete µ-map was generated where the water and fat were flipped in the DIXON sequences producing inaccurate µ-maps and reduced Standardize Uptake Values (SUV). The aim of this project was to validate the creation and application of µ-maps generated with only the internal body coil in breast PET/MR imaging using new MR considerations. Methods: A volunteer was imaged utilizing DIXON sequences with two different coil configurations, the breast plus body matrix coil (the source of truth) and the internal body coil only. Two different types of breast tissue were evaluated this year, one as water (saline) and a second as fat (vegetable oil). To simulate lesions in a patient, three conical 1.5 mL vials containing a clinically relevant concentration of F18-FDG were placed in the mediastinal region of the right breast and in the central regions of both the right and left breasts. SUVs were obtained from regions of interest drawn around all three simulated lesions in both types of breast tissue. These values were compared for each generated µ-map and PET reconstruction - breast and body matrix coils (complete) and internal body coil (complete). Results: Please see Table 1. Conclusions: The results revealed similar SUVmax measurements between the breast and body matrix coils and the internal body coil with a variance of less than 5% regardless of the type of simulated breast tissue. These results indicate that µ-maps created with the internal body coil are a viable alternative to those generated with the breast and body matrix coils. These results should be further assessed and verified with patient imaging rather than simulated breast tissue.