Dual Energy Radiography

Dual Energy Imaging with Dual Source CT Systems Rainer Raupach, PhD Siemens Healthcare [email protected] Radiograph

Bone image 2 energies

Tissue image

2 materials

Armato SG III. Experimental Lung Research. 2004;30 (suppl 1):72-77.

kV Switching with SOMATOM DRH – in the 80s Calculation of material selective images: Calcium and soft tissue

Principle of Dual Energy CT Data acquisition with different X-ray spectra: 80 kV / 140 kV

Standard image

Rapid kVp switching

Mean Energy: Calcium image

56 kV

76 kV

Basis material decomposition

Low kVp

Tube 1 Soft tissue image

Tube 2

High kVp

Attenuation profiles

Different mean energies of the X-ray quanta

W. Kalender: Vertebral Bone Mineral Analysis, Radiology 164:419-423 (1987)

1

SOMATOM Definition The World’s First Dual Source CT

Principle of Dual Energy CT Many materials show different attenuation at different mean energies

Faster than Every Beating Heart 1.0E+02

gated mode / same kV high temporal resolution (80ms) Cardiac imaging

Iodine Bone

Attenuation

56 kV 76 kV 1.0E+01 Large increase

One-Stop Diagnosis in Acute Care non gated mode / same kV low temporal resolution Obese patients, low kV scanning

1.0E+00 Small increase 1.0E-01 10

30

50

70 90 Energy / keV

110

130

Beyond Visualization with Dual Energy

150

different kV (gated and non-gated)

Reason: different attenuation mechanisms (Compton vs photo effect)

“Contrast Enhanced Viewing” using Dual Energy Information in Addition to Simple Image Mixing

Spectra of Dual Energy Applications

Basic application: Enhanced viewing, contrast optimization Contrast enhanced studies: Iodine has much higher contrast at 80 kV Non-linear, attenuation-dependent blending of the images combines benefits of 80 kV (high contrast) and mixed data (low noise) Direct Angio

Lung PBV

Virtual Unenhanced

Lung Vessels

140 kV

Hardplaque Display

Heart PBV

Musculoskeletal

Gout

Calculi Characterization

Lung Nodules* *510(k) approved

80 kV

Blending

Brain Hemorrhage

Xenon* Courtesy of CIC Mayo Clinic Rochester, MN, USA

2

syngo Dual Energy Direct subtraction of bone

syngo Dual Energy Direct subtraction of bone Modified 2-material decomposition: Separation of two materials Assume mixture of blood + iodine (unknown density) and bone marrow + bone (unknown density) Separation line 600

Iodine pixels

Bone 550 HU

HU at 80 kV

500 400

Automatic bone removal without user interaction Clinical benefits in complicated anatomical situations: Base of the skull Carotid arteries Vertebral arteries Peripheral runoffs

Bone pixels

Blood+iodine

80kV

Marrow+bone

300

Iodine 425 HU

Modified 2-material decomposition: Separation of bone and Iodine

200 100

Soft tissue

0 -100 -100

Bone 400 HU

Blood Marrow

Iodine 250 HU

140kV 0

100

200 300 HU at 140 kV

400

500

600

syngoDualEnergy Differentiation between hard plaques and contrast agent

Courtesy of Prof. Pasovic, University Hospital of Krakow, Poland

Image Based Methods Modified 2-material decomposition: Characterization of kidney stones Urine + calcified stones / uric acid stones

HU at 80 kV

high Z

low Z

HU at 140 kV

Courtesy of CCM Monaco, Monaco

3

syngo Dual Energy Musculoskeletal Visualization of tendons

syngo Dual Energy Visualization of Tendons: Tibialis posterior tendon rupture SOMATOM Definition World’s first DSCT Spatial Res. 0.33 mm Rotation 0.5 sec Scan time: 4 s Scan length: 133 mm 140/80 kV Eff mAs 80/150 Spiral Dual Energy

Courtesy of University Medical Center Grosshadern / Munich, Germany

Courtesy of University Medical Center Grosshadern / Munich, Germany

Applications of Dual Energy CT

Gout: Application

Three material decomposition: quantification of iodine – iodine image

HU at 80 kV

Iodine

Iodine content

65

Tissue

0

-100

Fat -90

0

60

HU at 140 kV

Vancouver General Hospital, Canada

Removal of iodine from the image: virtual non-contrast image

4

Image Based Methods

Applications of Dual Energy CT

Most promising application: 3-material decomposition Calculation of a virtual non-contrast image, Iodine quantification

Virtual non-contrast image and iodine image: Characterization of liver / kidney / lung tumors Solve ambiguity: low fat content or iodine-uptake Quantify iodine-uptake in the tumor and at the tumor surface Differentiation benign - malignant Monitoring of therapy response Mixed image

Mixed image 80kV+140kV

Virtual unenhanced image

VNC image

Iodine image

Iodine overlay image

+

Courtesy of University Hospital of Munich - Grosshadern / Munich, Germany

SOMATOM Definition Flash Latest Generation of Dual Energy CT

Applications of Dual Energy CT Quantification of iodine to visualize perfusion defects in the lung Avoids registration problems of non-dual energy subtraction methods 80/140kV Mixed Image

Iodine Image

Mixed image + iodine overlay

System Design Two X-ray tubes at 95°, each with 100 kW

33 cm

Two 128-slice detectors, each with 64x0.6mm collimation and z-flying focal spot Embolus

SFOV A/B-detector: 50/33 cm 0.28 s gantry rotation time 75 ms temporal resolution

Courtesy of Prof. J and M Remy, Hopital Calmette, Lille, France

5

Dual Energy Imaging with Tin Filtration ‘Definition’ vs. ‘Definition Flash’: Improved DE Signal

SOMATOM Definition Flash Single dose Dual Energy

Mixed Images

80 kV 140 kV overlap

VNC

Iodine

DE Images

Definition

Conventional DE

DSCT Dual Energy DE with Selective Photon Shield

80 kV 140 kV with SPS overlap

Tissue characterization Improved DE contrast Dose-neutral compared to a single 120 kV scan

Definition Flash

DE with Selective Photon Shield

SD and dose: equal

SD: -25%

Images acquired and processed in collaboration with CIC Mayo Clinic Rochester, USA

SOMATOM Definition Flash Impact of the Selective Photon Shield

SOMATOM Definition Flash Image Dual Energy Whole Body CTA: 100/140Sn kV @ 0.6mm

Dose neutral DE: comparison of 120 kV and 100 kV/140 kV+0.4 mm Sn Single DE CT Scan

120kV, 500mA

100/140Sn kV, 500mA

noise: 14.1 HU

noise: 13.9 HU

iodine: 329.0 HU

iodine: 330.0 HU

bone: 334.8 HU

bone: 335.3 HU

Courtesy of Friedrich-Alexander University Erlangen-Nuremberg - Institute of Medical Physics / Erlangen, Germany

6

Dual Energy CT

New Application Classes

Are there alternative approaches? 40 keV

Sequential acquisition at 80 kV and 140 kV with single source CT Registration problems (heart/lung motion, varying contrast density) Fast kVp-switching during the scan with single source CT Inadequate power at low kV Unequal noise for low and high kV data Measurement of Lung Nodule enhancement courtesy of ASAN Medical Center, Seoul, Korea

Measurement of Xenon Concentration

190 keV

Spectral sensitive „sandwich“ detectors Inferior spectral separation

courtesy of ASAN Medical Center, Seoul, Korea

Mono-energetic imaging courtesy of Klinikum Großhadern, Munich, Germany

Dual Energy CT Evaluation of alternative approaches

Quantum counting Paralysis at high flux rate Spectral overlap by fluorescence and pile-up

Dual Energy CT Evaluation of alternative approaches

dual−source (tin filter) dual−source (std. filter) sequential kVp dual−layer (GOS) dual−layer (CsI) dual−layer (ZnSe) quantum counting (CZT)

1.6 1.4

DE Performance @ equal dose

relative DEC²

1.2 1 0.8 0.6 0.4

Dose

0.2 0 15

20

25 30 35 phantom diameter [cm]

40

45

S. Kappler et al., Dual-energy performance of dual-kVp in comparison to dual-layer and quantum-counting CT system concepts, Proceedings of the SPIE Medical Imaging Conference, Volume 7258, pp. 725842 (2009)

7

Thank you!

8