Preclinical evaluation of MR-attenuation correction versus CT-attenuation correction on - PowerPoint Presentation

Slide1

Preclinical evaluation of MR-attenuation correction versus CT-attenuation correction on the sequential whole-body MR/PET

Jason Bini1,2, Jesus Mateo1,3, Josef Machac4, Jagat Narula5, Valentin Fuster3,5, Zahi Fayad1,5,6 , David Izquierdo-Garcia1 1 Translational and Molecular Imaging Institute, Mount Sinai School of Medicine, New York, NY, USA2 The City College of New York, Department of Biomedical Engineering, New York, NY, USA 3 The Centro Nacional de Investigaciones Cardiovasculares (CNIC), Madrid Spain4 Division of Nuclear Medicine, Department of Radiology, Mount Sinai School of Medicine., New York, NY, USA5 Department of Cardiology, Zena and Michael A Weiner Cardiovascular Institute and Marie-Josee and Henry R Kravis Cardiovascular Health Center., Mount Sinai School of Medicine, New York, NY, USA6 Department of Radiology, Mount Sinai School of Medicine, New York, NY, USA

Positron Emission

Tomography (PET)Functional imaging of physiologyHighly sensitive probes (10-12 Molar)Low spatial and anatomical resolutionAnatomical imaging from CT scanHigh spatial resolution anatomic imagingFunctional imaging possible but relatively low probe detection sensitivity (10-6 Molar)Magnetic Resonance Imaging (MRI)MR/PET

Many technical challenges to combine PET and MR

MR provides superior soft tissue contrast vs. CTMR/PET lower radiation dose to patient vs. PET/CTMR/PETCombined Sequential Whole-bodyPhilips Ingenuity TF sequential whole-body PET/MR system3T = 30000G3T MRI10G3m

PET

Attenuation Correction

CT to PET 511keV attenuation coefficient conversionPET/CT attenuation correction (AC)Clinical gold standardLow-dose CT scans easily transformed to attenuation map (CTAC) at 511 keV for use in PET reconstruction algorithms 1MRINo direct information about photon attenuationMeasures proton densities and magnetic relaxation timesMR-based AC (MRAC)Non-trivial method of assigning MR signal intensity to empircial photon attenuation coefficients in each voxel1 Burger C, et al. Eur J Nuc Med Mol Img. 2002;29(7):922–7.

Attenuation Correction

MR attenuation mapsCT attenuation mapsAttenuation correction implemented on Philips sequential whole-body MR/PETT1-wieghted MR acquisition that matches PET dimensionsSimilar to low-dose CT in PET/CTThree class segmentation algorithm following human-like a priori model (air, lungs and soft tissue)Animal imaging Lungs fail to segment - rabbit lung volumes do not match human-like a priori model2-segment classification (soft tissue and air)Soft tissue (0.095) 1,2 Air (0)

1

Meikle

SR,

et al

. JNM. 1993;34(1):143–50.

2

Bettinardi

V,

et al

.

Eur

J

Nuc

Med Mol

Img

. 1999;26(5):447–58.

MRAC vs. CTAC

Objective: Evaluate MRAC implemented on the combined sequential MR/PET scanner versus CTAC with the same PET data in an animal model

Animal Model

20 atherosclerotic New Zealand White male rabbits (3.8±0.3kg)Retrospectively analyzed 4-5 mCi injected of either… Fluoromisonidazole (18F-FMISO) (n=12) Fluorodeoxyglucose (18F-FDG) (n=8) CT acquisition Non-contrast low-dose CT Stand-alone CT scanner Philips Brilliance iCT -- 256 slice multidetector CT Voxel size 0.625x0.625x1.0mm Image matrix 512x512x399CT Imaging Protocol

PET Imaging Protocol

PET acquisition 3D mode Time of flight1 bed position/15 minutes – 3 hours post injection Images reconstructed into 128x128x90 matrix Voxel size 2x2x2mm 3D RAMLA reconstruction with 3 iterations and 33 subsets Philips system standard corrections Normalization, dead time, decay, scatter, random coincidences, sensitivity Attenuation correction provided by MRAC or CTAC methods No MR coils on table MR/PET table template inserted in reconstruction

PET Image Reconstruction

CTMRI CT and MR now both in PET spaceMR attenuation mapCT attenuation map - bilinear conversion to 511keV attenuation coefficients PET reconstruction

PET

CTAC

PET

MRAC

CT

coregsitered

to MR

MR attenuation map

CT attenuation mapPET with CTACPET with MRACPercent difference mapPercent difference map

PET quantification

voxel-by-voxelAverage difference for all voxels-0.94% (-0.06±0.30SD)R^2=0.99, p<0.0001

PET quantification

Regions-of-interest (ROIs)Six ROIsAorta, liver, left and right kidneys, spine and soft tissue (back muscles)Mean SUV ± Standard Deviation (SUVmin, SUVmax)ROIPET CTAC

PET

MRAC

p

AORTA

0.30

±0.22

(0.25, 0.37)

0.27

±0.22 (0.22, 0.35)

<0.0001

LIVER

1.06

±

0.68 (0.54, 1.40)

0.95

±

0.62 (0.47, 1.27)

<0.0001

LEFT KIDNEY

1.24

±

0.55 (0.34, 3.02)

1.12

±

0.48 (0.31, 2.47)

<0.0001

RIGHT KIDNEY

1.22

±

0.56 (0.31, 3.11)

1.14

±0.53 (0.30, 2.91)

<0.0001

SPINE

0.42

±

0.30 (0.21, 0.71)

0.31

±

0.25 (0.15, 0.54)

<0.0001

SOFT TISSUE

0.19

±

0.11 (0.13, 0.45)

0.16

±

0.10 (0.11, 0.38)

<0.0001

-9.7% (-0.15±0.12SD)

R^2=0.99, p<0.0001-10.8% (-0.08±0.06SD)R^2=0.99, p<0.0001PET quantificationRegions-of-interest (ROIs)

PET quantification

Regions-of-interest (ROIs)RabbitsPercent Difference SUVmeanPercent Difference SUVmaxALL

ROIs

ALL

-10.8

-9.7

FDG

-10.6

-10.4

FMISO

-11.0

-9.2

Aorta

ALL

-10.4

-7.6

FDG

-4.3

-2.6

FMISO

-21.2

-17.4

Liver

ALL

-10.4

-9.3

FDG

-9.5

-8.7

FMISO

-12.1

-10.3

Left

Kidney

ALL

-9.8

-9.5

FDG

-10.2

-10.1

FMISO

-9.5

-9.1

Right

Kidney

ALL

-6.2

-6.4

FDG

-7.2

-7.5

FMISO

-5.5

-5.8

Spine

ALL

-26.1

-23.6

FDG

-21.7

-19.5

FMISO

-35.5

-32.2

Soft

Tissue

ALL

-16.8

-15.5

FDG

-17.7

-17.4

FMISO

-15.8

-12.5

Rabbits

Percent Difference

SUVmean

Percent Difference

SUVmax

PET quantification

Regions-of-interest (ROIs)RabbitsAverageSUVmean

Absolute Difference

SUVmean

PET

CTAC

PET

MRAC

Left

Kidney

ALL

1.24

1.12

-0.12

FDG

1.26

1.13

-0.13

FMISO

1.23

1.12

-0.11

Right

Kidney

ALL

1.22

1.14

-0.08

FDG

1.16

1.08

-0.08

FMISO

1.25

1.19

-0.06

Spine

ALL

0.42

0.31

-0.11

FDG

0.72

0.57

-0.15

FMISO

0.22

0.14

-0.08

Soft

Tissue

ALL

0.20

0.16

-0.04

FDG

0.25

0.20

-0.05

FMISO

0.16

0.14

-0.02

Rabbits

Average

SUVmean

Absolute

Difference

SUVmean

PET

CTAC

PET

MRAC

ALL

ROIs

ALL

0.74

0.66

-0.08

FDG

0.94

0.84

-0.10

FMISO

0.61

0.54

-0.07

Aorta

ALL

0.30

0.27

-0.03

FDG

0.49

0.47

-0.02

FMISO

0.18

0.15

-0.03

Liver

ALL

1.07

0.95

-0.12

FDG

1.76

1.59

-0.17

FMISO

0.60

0.53

-0.07

Martinez-Möller

, et al 1Osseous lesions underestimated by 8.0% without bone segmentationHuman study - fat, soft-tissue, lungs and air segmentationPET/CT – CT map bone set to soft tissue attenuation coefficients of 0.10Hoffman, et al 2PET/CT with bone set to soft tissue (no reported soft tissue attenuation coefficient)Underestimation of 60% in spineKeereman, et al 3Human Whole-body MR/PET simulation studyTwo lesions in the spine underestimated by 11.7% and 10.8% (spongeous bone -> soft tissue) Same lesions underestimated by 15.6 and 17.5% (cortical bone ->soft tissue)Lesion in the pelvis underestimated by 13.1% (sponegous bone -> soft tissue)Soft tissue attenuation coefficient of 0.0968Steinberg, et al 4Underestimation in the spine of 14%Three segment classification, lung, soft tissue and air in beagles

Segmented MR with soft tissue attenuation coefficients of 0.098 coregistered

to PET/CT

Hu

,

et al

5

Human, three-region segmentation implemented on

Philips MR/PET scanner (15 patients)

Segmented MR

coregistered

with PET/CT

Underestimated lesions by 7.6% in spine and 11.7% in pelvis

In the context of the literature…

3

Keereman

V,

et al

. Med Phys. 2011;38(11):6010-9.

1

Martinez-

Möller

,

et al

. Journal of Nuclear Medicine. 2009;50(4):520–6

.

4

Steinberg J,

et al

. Nuclear medicine and biology. 2010;37(2):227–35.

5

Hu

Z,

et al

. Conference Record (NSS/MIC). 2009. p. 3508–12.

2

Hofmann M, et al.

Eur

J of

Nuc

Med Mol

Img

. 2009;36

Suppl

1:S93–104.

Wide range of underestimation differences in literature when ignoring bone (7 to 60%)

Results suggest the need for bone in MR attenuation maps to minimize quantification errorsAnimal model may impact degree of underestimation when bone is replaced with soft tissue due to percentage of total body volume that represents the skeleton1,2Ultrashort Echo Time (UTE) sequences for head and neck imagingNot solved for whole-body imagingLong acquisition timeDixon sequences to distinguish fat and soft tissue have been implemented for abdomen, head and neckLeft and right kidney discrepancyPETMRAC2 Keereman V, et al. Med Phys. 2011;38(11):6010-9. 1Steinberg J, et al. Nuclear medicine and biology. 2010;37(2):227–35.

We have confirmed the MR/PET sequential scanner underestimates PET values by <10% in most regions

Areas in or close to bone(spine and back muscles) underestimate values by >10%Despite underestimation, in and near bone, two segment classification of air and soft tissue for preclinical MRAC provides reasonable SUV quantification for sequential MR/PETPETMRAC

Acknowledgements

Cardio-Image Program from Centro Nacional de Investigaciones Cardiovasculares (CNIC), Spain (J.M., V.F.)Partial support was provided by NIH/NHLBI R01 HL071021, R01 HL078667 and NIH/NCRR CTSA UL1RR029887 (Imaging Core) (Z.A.F)