New Advances in CT: Functional Imaging & Dose Reduction - - PowerPoint PPT Presentation

▶
new advances in ct functional imaging amp dose reduction
SMART_READER_LITE
LIVE PREVIEW

New Advances in CT: Functional Imaging & Dose Reduction - - PowerPoint PPT Presentation

Aug 07, 2023 259 likes •731 views

New Advances in CT: Functional Imaging & Dose Reduction Ting-Yim Lee PhD, FCCPM; Xiaogang Chen PhD; Eugene Wong PhD; Nikolaj Jensen BSc; Glenn Bauman MD; Michael Lock MD; Barbara Fisher MD Lawson Health Research Institute; Robarts Research

slide-1
SLIDE 1

New Advances in CT: Functional Imaging & Dose Reduction

Ting-Yim Lee PhD, FCCPM; Xiaogang Chen PhD; Eugene Wong PhD; Nikolaj Jensen BSc; Glenn Bauman MD; Michael Lock MD; Barbara Fisher MD

Lawson Health Research Institute; Robarts Research Institute; Medical Imaging, Physics & Astronomy and Oncology, The University of Western Ontario; London Regional Cancer Program London, Ontario, CANADA

slide-2
SLIDE 2

Functional Imaging in Radiation Oncology

  • Target definition

– Areas of increased invasiveness and resistance

  • Enhanced angiogenesis
  • Hypoxia

– GTV, CTV

  • Monitoring of tumor response to treatment

– Tumor mass – Reoxygenation – Shut down of vasculature – Adjuvant chemo and anti-angiogenesis therapy

  • Vascular normalization**

**Jain. Science 307:58, 2005

slide-3
SLIDE 3

Functional Imaging Modalities

  • Positron Emission Tomography (PET)
  • Single Photon Emission Tomography (SPECT)
  • Magnetic Resonance Imaging (MRI)
  • Computed Tomography ?

– Anatomical/Morphological

  • Introduce CT Functional Imaging

– Brain and liver tumor

  • Radiation dose from CT

– Methods to reduce dose – In particular for CT functional imaging

slide-4
SLIDE 4

CT Perfusion Brain Tumor Study

Intravenous Injection of Contrast Agent 40-50 ml @ 3- 4 ml/s

CBF ml⋅min-1⋅(100g)-1

Scan Protocol

  • 80 kVp, 190mA, 4 x 5

mm

  • 45 s cine scan
  • 1 s every 16 s, 6 times

Deconvolution

MTT (s) CBV ml⋅(100g)-1 PS ml⋅min-1⋅(100g)-1 200 20 12 12

  • 50

50 100 150 200 250 300 350 400 40 80 120 160

Time (s) HU

Artery Vein

  • 5

5 10 15 20 25 30 35 40 80 120 160

Time (s) HU

Tumor Normal Brain

  • 5

5 10 15 20 25 30 35 40 80 120 160

Time (s) HU

Tumor Normal Brain

slide-5
SLIDE 5

Basis of CT Perfusion (Functional) Imaging

  • Tissue curve is dependent on

– Blood flow, blood volume, mean transit time, capillary permeability surface area product (PS) – Arterial input concentration

  • Tracer kinetics modeling

– Models tissue curve with above parameters and arterial concentration

  • Fitting of measured tissue

curve with model curve

– Estimates of parameters

  • 50

50 100 150 200 40 80 120 160

Time (s) HU

Artery

  • 5

5 10 15 20 25 30 35 40 80 120 160

Time (s) HU

Tumor Fitted Curve

  • perator

n Convolutio : (IRF) Function Residue Impulse : R(t) (CBF) flow blood Cerebral : F curve Arterial : ) t ( C curve brain

  • r

Tumor : ) t ( Q ) t ( R ) t ( C F ) t ( Q

a a

∗ ∗ ⋅ =

slide-6
SLIDE 6

Modeling of Tissue Contrast Kinetics

  • Johnson and Wilson Model*

– Blood flow, transit time, transcapillary exchange, ven. outflow – Exchange regimes

  • Diffusion limited (F << PS); cold xenon
  • Flow limited (F>>PS); intravascular tracer: contrast agent in normal

brain

  • Permeable (F ~ PS); contrast agent in most tissues, e.g. liver

FCa (t) FCv (t) PS EES, Ce (t), ve x Blood Space, Cb (x,t), vb

*Am J Physiol

210:1299-1303, 1966

slide-7
SLIDE 7
  • Tissue Blood Flow (CBF, F)
  • Tissue Blood Volume (CBV, vb

)

  • Microvessel

Permeability Surface Area (PS)

  • Mean transit time (MTT, Tc

)

FCv(t) PS ≅ FE EES, Ce (t), ve x Blood Space, Cb (x,t), vb

Q(t)

FCa (t)

Tc R(t) Time (1-E) vb PS = - F ln(1-E) k = FE/ve E·e-kt E 1.0

CT Perfusion Imaging in Brain Tumors

) t ( R ) t ( C F ) t ( Q

a

∗ ⋅ =

Lee: Q J Nucl Med 41:171-87, 2003

slide-8
SLIDE 8

Microspheres CBF (ml/min/100g)

50 100 150 200 250

CT CBF (ml/min/100g)

50 100 150 200 250

Tumour Peri-Tumour Contra-lateral Normal

CT vs. Microspheres CBF Measurements in Brain Tumours

RA

Validation : Accuracy and Reproducibility

  • Rabbit VX2 brain tumor model
  • 13% and 7% for CBF and CBV

m=0.97 r=0.86

Cenic & Lee: AJNR 21:462-470, 2000

slide-9
SLIDE 9

Brain Tumor Studies

  • Primary brain tumor – glioma
  • Before and 1-2 weeks post radiation therapy
  • Changes in CT Perfusion functional parameters

and response to treatment

slide-10
SLIDE 10

Glioma-2

slide-11
SLIDE 11
  • Blood flow map fused with planning CT
  • Whole brain irradiation
  • 50Gy/25

Glioma-2

slide-12
SLIDE 12

Glioma-2

Before After

CECT Blood Flow Blood Volume MTT PS

slide-13
SLIDE 13

Glioma-2

Before After

CECT Blood Flow Blood Volume MTT PS

slide-14
SLIDE 14

Tumor Regions

30 60 90 120 150 BF (ml/min/100g) BV x 10 (ml/100g) PS x 100 (ml/min/100g) MTT X 10 (s) Pre-Treatment Post-Treatment

∗∗ P < 0.05 ∗∗ ∗∗

Glioma-2

slide-15
SLIDE 15

Normal Regions

15 30 45 60 BF (ml/min/100g) BV x 10 (ml/100g) PS x 100 (ml/min/100g) MTT X 10 (s) Pre-Treatment Post-Treatment

∗∗ P < 0.05 ∗∗ ∗∗

Glioma-2

slide-16
SLIDE 16

Mar 23 2006 Jun 21 2006 Sep 19 2006 Mar 07 2007

Glioma-2

slide-17
SLIDE 17
  • 50

50 100 150 200 250 20 40 60 80 100 120

Time (s) HU

Aorta Portal Vein

CT Liver Perfusion Study

Intravenous Injection of Contrast Agent

BF BV MTT PS HAF

Axial Shuttle

Hepatic Artery PortaVein Hepatic Vein

Deconvolution

  • 10

10 20 30 40 20 40 60 80 100 120

Time (s) HU Tumor Normal Liver

  • 50

50 100 150 200 250 20 40 60 80 100 120

Time (s) HU

Aorta Portal Vein

[ ]

(IRF) Function Residue Impulse : R(t) (HAF) fraction arterial Hepatic : α (BF) flow blood hepatic Total : F curve venous Portal : ) t ( C curve Aortic : ) t ( C curve liver

  • r

Tumor : ) t ( Q ) t ( R ) t ( C ) α 1 ( ) t ( C α F ) t ( Q

p a p a

∗ ⋅ − + ⋅ ⋅ =

slide-18
SLIDE 18

CT Liver Perfusion Study

  • Couch ‘shuttles’

free breathing patient between two locations

  • Each location is scanned

every 2.8 s

  • 8 x 5 mm thick slices for each

location

  • Total 80 mm coverage with 16

x 5 mm slices

  • Images have to be sorted to

reduce patient motion

slide-19
SLIDE 19

Unsorted Sorted

  • 50

50 100 150 200 250 300 20 40 60 80 100 120

Time (s) HU

Aorta Portal Vein

  • 50

50 100 150 200 250 300 20 40 60 80 100 120

Time (s) HU

Aorta Portal Vein

CT Liver Perfusion Study

slide-20
SLIDE 20

Total Liver Blood Flow

(ml⋅min-1⋅(100g)-1)

CT Liver Functional Maps

Hepatic Arterial Fraction (HAF)

(% Total Blood Flow from Artery)

  • Metastasis was more hypovascular

than surround normal liver tissue

  • Does it mean that the metastasis was hypoxic ?
slide-21
SLIDE 21

Hepatic Arterial Blood Flow

(ml⋅min-1⋅(100g)-1) Total Blood Flow HAF

CT Liver Functional Maps

  • Hepatic arterial blood flow (HA

BF) in metastasis is greater than or equal to that in normal liver tissue

  • Metastasis might not be hypoxic, particularly in the rim
  • Indicates feasibility of radiation treatment for liver tumor?
slide-22
SLIDE 22

Validation of CT Liver Perfusion

  • Seven New Zealand White Rabbits with implanted VX2

tumor in the liver

Stewart & Lee. PMB 53:4249, 2008

slide-23
SLIDE 23

CT Liver Perfusion Study

  • Objective

– CT Perfusion functional parameters to distinguish tumor from normal tissue

  • 7 Hepatocellular

Carcinoma (HCC), 5 metastases and 1 cholangioma

  • Each subject had a CT Perfusion Liver study with

the free breathing axial shuttle technique

  • 120 kVp, 60 mAs, 0.4 s rotation period, 42 passes
  • f the axial shuttle with the first 4 as baseline
  • 60 –

70 ml of Omnipaque 300 injected at 3 ml⋅s-1 at the 5th pass of the axial shuttle

slide-24
SLIDE 24

10 20 30 40 50 60 70

Tumor Nomal

Hepatic Arterial Fraction (%) 20 40 60 80 100 120 140 160

Tumor Nomal

Total Blood Flow

(ml/min/100g)

10 20 30 40 50 60 70 80

Tumor Nomal

Arterial Blood Flow

(ml/min/100g)

CT Liver Perfusion Study

  • Results

– Total blood flow was lower while arterial blood flow was higher in the tumor than in normal tissue – Tumor may not be hypoxic

P < 0.05 P < 0.05 P < 0.05

slide-25
SLIDE 25

CT Radiation Dose

  • NCRP Report 160 –

Ionizing Radiation Exposure of the Population of the US

– 2006

  • 414 million X-ray procedures

– 62 million CTs (15%)

  • Collective radiation dose

– 900,000** person-Sv (vs 124,000 in 1990) – 918,000 person-Sv from background – 440,000 person-Sv from CT

**

From all ionizing radiation procedures including Nuclear Medicine and therapy

NEJM 357:2277, 2007

slide-26
SLIDE 26

CT Radiation Dose

  • Typical effective dose of CT studies

**Shrimpton

et al. NRPB-W67, 2005 CT Scan UK 2003** Europe 2004**

DLP (mGy-cm) Effective Dose (mSv) DLP (mGy-cm) Effective Dose (mSv) Routine Head 931 2.0 989 2.1 Abdomen (Liver metastases) 472 7.1 989 14.8 Chest, abdomen & pelvis (lymphoma staging or follow up 937 14.1

  • Chest (lung cancer: known,

suspected or metastases) 575 8.1 430 6.0

CT Functional Study

DLP (mGy-cm) Effective Dose (mSv) Head (4 cm) 2289 4.8 Liver (axial shuttle – 8 cm) 1268 19.0

slide-27
SLIDE 27

CT Dose Reduction – mA Modulation

MHRA Rep 05016, 2005, UK

a) with patient size (adult vs children) b) according to patient thick-ness along longitudinal (z) axis c) according to patient thickness in the transaxial plane d) Combination of a), b) and c) e) Dose shield for helical scan

slide-28
SLIDE 28

CT Dose Reduction

Kalra et al. Radiology 233:649, 2004

  • mA

modulation in the transaxial plane

  • Dose reduction 22 –

34%**

**Greess

et al. Eur Radiol 12:1571, 2002 Mastora et al. Eur Radiol 11:590, 2001

slide-29
SLIDE 29

CT Dose Reduction

Kalra et al. Radiology 233:649, 2004; **232:347, 2004

  • mA

modulation along longitudinal (z) axis

  • Dose reduction by

32% on average**

slide-30
SLIDE 30

CT Dose Reduction

Carrington; Somaton Sessions 22:6, 2008

  • Beam collimation to reduce overscan

in axial direction

  • Dose reduction by 13-36%**

**Tzedakis et al. Med Phys 32:1621, 2005

slide-31
SLIDE 31

CT Dose Reduction

Thibault et al. Med Phys 34:4526, 2007

  • Adaptive Statistical Iterative Reconstruction (ASIR)
slide-32
SLIDE 32

Dose Reduction by ASIR

  • Liver scan –

portal phase

w/o ASIR

DLP: 836 mGy-cm

w ASIR

DLP: 688 mGy-cm 18% less

slide-33
SLIDE 33

Dose Reduction by ASIR

  • Brain scan

w/o ASIR

DLP: 1018 mGy-cm

w ASIR

DLP: 939 mGy-cm 8% less

slide-34
SLIDE 34

Dose Reduction by ASIR

80 kVp 200 mAs 80 kVp 50 mAs 80 kVp 50 mAs + ASIR

  • Brain Functional Study

– 4 patients recruited – Each patient had a high dose and low dose CT Perfusion studies separated by 10 min – High and low dose study used 200 and 50 mAs per image – The low dose study was also processed by ASIR

slide-35
SLIDE 35

Dose Reduction by ASIR

  • Brain Functional Study: Average Map

80 kVp 200 mAs 80 kVp 50 mAs 80 kVp 50 mAs + ASIR

slide-36
SLIDE 36

Dose Reduction by ASIR

  • Brain Functional Study

80 kVp 200 mAs 80 kVp 50 mAs 80 kVp 50 mAs + ASIR

Blood Flow Blood Volume

slide-37
SLIDE 37

Dose Reduction by ASIR

  • Figure of Merit (FOM) of Functional Maps

– Segment out grey and white matter – Grey and white matter pixel mask on functional maps – Mean (μ) and standard deviation (σ)

  • f functional parameter

– μ σ FOM =

slide-38
SLIDE 38

Dose Reduction by ASIR

  • Figure of Merit (FOM)

0.1 0.2 0.3 0.4 0.5 0.6 0.7 High Dose Low Dose LD+ASIR

White Matter Blood Flow

0.1 0.2 0.3 0.4 0.5 0.6 0.7 High Dose Low Dose LD+ASIR

White Matter Blood Volume

P < 0.05 P < 0.05

slide-39
SLIDE 39

Dose Reduction by ASIR

  • Figure of Merit (FOM)

0.1 0.2 0.3 0.4 0.5 0.6 High Dose Low Dose LD+ASIR

Grey Matter Blood Flow

0.1 0.2 0.3 0.4 0.5 0.6 High Dose Low Dose LD+ASIR

Grey Matter Blood Volume

P < 0.05 P < 0.05

slide-40
SLIDE 40

Comparison of Functional Imaging Modalities

Features MR CT PET SPECT

Speed

√√ √√√ √√ √

Resolution

√√ √√√ √√ √

Axial coverage

√√ √ √√ √√

Radiation dose no xx x x On-site cyclotron no no yes no Quantification

√√ √√√ √√√ √

Cost x x x x x x x x x

slide-41
SLIDE 41

Conclusion

  • CT Perfusion (Functional) imaging is an easy

extension of routine contrast enhanced CT scanning

  • Based on ‘realistic’

modeling of transport kinetics of blood-borne contrast agents

  • Blood flow is validated against microspheres
  • Define targets for radiation treatment
  • Monitor progress and treatment response of

tumors

  • With new radiation dose reduction

techniques, viable alternative to PET and MR

slide-42
SLIDE 42

Dose Reduction by ASIR

80 kVp 200 mAs 80 kVp 20 mAs 80 kVp 20 mAs + ASIR

  • Brain Functional Study

– 4 patients recruited – Each patient had a high dose and low dose CT Perfusion studies separated by 10 min – High and low dose study used 200 and 50 mAs per image – The low dose study was also processed by ASIR

slide-43
SLIDE 43

Dose Reduction by ASIR

  • Brain Functional Study: Average Map

80 kVp 200 mAs 80 kVp 20 mAs 80 kVp 20 mAs + ASIR

slide-44
SLIDE 44

Dose Reduction by ASIR

  • Brain Functional Study

80 kVp 200 mAs 80 kVp 20 mAs 80 kVp 20 mAs + ASIR

Blood Flow Blood Volume

slide-45
SLIDE 45

Dose Reduction by ASIR

  • Figure of Merit (FOM) of Functional Maps

– Segment out grey and white matter – Grey and white matter pixel mask on functional maps – Mean (μ) and standard deviation (σ)

  • f functional parameter

– μ σ FOM =

slide-46
SLIDE 46

Acknowledgement

My Lab: A Cenic A So DG Nabavi E Stewart E Henderson G Carnes T Purdie C d’Esterre S Gaede M Dekaban A Laslo P Gabra B Murphy X Chen JF Adam J Hadway D Ouimet L Desjardins

M Murphy F Su J Bonasia

Collaborators: R Kozak J Amann CT Staff at SJHC