MR-based Attenuation Correction for PET/MR - Correction Techniques in Emission Tomography - page 220

bined devices were not yet considered seriously for clinical use. These early
studies focused on PET applications in neurology [14, 41]. With the considera-
tions of clinical PET/MR prototypes several groups have proposed algorithms
for extra-cranial MR-AC as well [3, 15].
MR-AC can be evaluated by comparing the PET images that are obtained
following standard TX-AC (transmission scan-based attenuation correction)
or CT-AC and MR-AC. In the following we will refer to these images as
PET TXAC , PET CTAC and PET MRAC , respectively. Both TX-AC and CT-AC
have their shortcomings. Depending on the scan time, TX scans exhibit rela-
tively high noise levels that can be detrimental for AC purposes. On the other
hand, CT images have noise levels that are several orders of magnitude lower
than those obtained from images acquired with standard transmission sources
(TX). However, the mapping of CT-based Hounsfield units to 511 keV attenu-
ation values can be incorrect, particularly in the case of non-organic materials
such as metal implants. Despite these problems, both TX-AC and CT-AC
are commonly used and accepted. In accordance with the literature we will
present PET TXAC or PET CTAC as the gold standard against which PET MRAC
is compared. The comparison can be done visually or quantitatively by means
of relative differences of the reconstructed PET activity distributions. Differ-
ences can be assessed on a voxel-by-voxel basis, or, perhaps, more commonly
for regions-of-interest (ROI).
ROIs can be defined automatically or manually by a clinical expert. For a
study with n patients, where p ROIs are defined for each patient, it is impracti-
cal to quote all n p dierences. Therefore, it is preferred to report either the
maximum differences or the mean absolute difference across all voxels. Some
authors have quoted the mean differences, where the mean was taken from
the positive or negative differences. This value indicates only the existence of
an overall bias in the method, a value of zero for the differences would merely
indicate that activity was overestimated as often as it was underestimated.
Table 11.1 summarizes the results of the most significant studies on MR-AC.
11.2 MR-AC for brain applications
11.2.1 Segmentation approaches
MR-AC for brain applications was first addressed by Le Goff-Rougetet et
al., who proposed a method to calculate PET AC factors from MR images
in clinical examinations when both PET and MRI were required [14]. They
argued that MR-AC helps simplify the clinical protocol and reduce the pa-
tient dose from standard PET transmission scanning. The methodology of
their work, which they first applied to an FDG/water-filled cylindrical Lucite
phantom, is based on a co-registration of the MR images to the PET trans-