Transcript 2016_04_12_DetOpt_aortegax

Work progress report from
UC3M
Ana Ortega-Gil
Juan José Vaquero
Supported by the INFIERI EU project n°317446, FP7-PEOPLE-2012-ITN
Pleased to be here
Bio
Ph.D. Student in Electrical Engineering,
Electronics and Automation program at UC3M
(2015-2018)
“Optimization of micro-CT imaging
systems for soft-tissue dedicated
tasks”
Last January, I joined INFIERI Young Research Panel for representing
the non EU-funded fellows in the INFIERI network
2
Outline
1. Context and goals of our work
2. Materials
3. Methods: System optimization
o
o
o
Spectral configuration
Detector response
Geometrical configuration
4. Results
o
Optimized protocol
5. Towards Spectral CT
3
Goals of our work
Optimize micro computed tomography systems

Lung damage quantification based on the intensity
differences between healthy and sick lung


Assess the progression of the disease
Help to monitor the response to treatment
4
CT Imaging technique
Structural
modality
medical
that
imaging
provides
a
volumetric representation of the
attenuation coefficient shown
by the different body tissues to
the incoming x-ray radiation
Y
Absolute coordinates
Z
X
5
In-vitro CT Imaging technique
Structural
modality
medical
that
imaging
provides
a
volumetric representation of the
attenuation coefficient shown
by the different body tissues to
the incoming x-ray radiation
Y
Absolute coordinates
Z
X
6
Hardware
 Testbench CT:
•
•
High-resolution in-vitro x-ray micro-CT
(UC3M designed)
• Applications in research and teaching: freedom on
geometry and component choice
Transferred to Sedecal S.L for assembling
(framed outside INFIERI)
7
Hardware
 Testbench CT:
•
•
High-resolution in-vitro x-ray micro-CT
(UC3M designed)
• Applications in research and teaching: freedom on
geometry design and modules
Transferred to Sedecal S.L for assembling
(framed in IMI-Predict Tb)
8
Hardware
 Testbench CT:
Microfocus X-ray source: L9631
Max.Power
50 W
Voltage
40 - 110 kVp
Focal spot
20 – 130 um (6 W – 50 W)
Hi-res CMOS based flat-panel detector
Pixel size
Eff. area
#pixels
0.075 (Binning 1x1)
150 x 122 mm
1944x1536 pixels
Frame rate
26-86 fps (Binning 1x1 - 4x4)
Scintillator
0.15 mm-thick CsI:TI
Mechanical system
5 linear
1 rotatory
ISCM-4805
9
System optimization
Optimization and calibration*
Optimization

Working point with
0.2mm Cu filtration

Trade off magnificationenergy
Acquisition
Imaging
protocol
Reconstruction
* based on acquisition protocol optimization tasks by
Sisniega A. et al, “Monte Carlo Study of the Effects of System Geometry and Antyscatter Grids on CBCT
Scatter Distributions ”, Med.Phys, 40,051915 2013
Sisniega A. et al, “Design And Assessment Principles of Semiconductor Flat-panel Detector-based X-Ray MicroCT Systems for Small Animal Imaging “ in K.Iniewski, Ed.,ed New Yotk CRC Press, 309-335,2012
10
System optimization
 We characterize the system as end users by:
Optimization and calibration
Optimization

Working point with
0.2mm Cu filtration

Trade off magnificationenergy
•Acquisition protocol
•Geometrical parameters
11
Spectral configuration
System optimization
OBJECTIVE: To select voltage, current and filtration
for soft-tissue dedicated task
METHOD
 Working point
 Compare the signal provided by the system detector to
an ideal detector with
 Flat energy response
 Pixel gain equal to the average gain our detector.
EFFECT ON IMAGE QUALITY
Voltage (kV) and filtration-> penetration -> contrast
Current (uA) -> saturation
12
Detector Response
System optimization
OBJECTIVE: Detector general response as a function
of exposure level, working point selection.
METHOD
 Check the linearity of the pixel signal values
 Mean of individual pixel response and of individual
variances
 Select the maximum source-to-detector signal: signal
level at detector of 90% the saturation one.
 Stability estimated as
 Mean pixel signal over a given period of time for fixed
settings (standard acquisition protocol)
EFFECT ON IMAGE QUALITY
Intensity inhomogeneity within tissue
13
Optimal Acquisition Protocol
Soft Tissue (optimized) protocol
Mouse
68 kV (+ 0.2 mm Cu)
Conventional protocol
Mouse
50 kV (+ 1.0 mm Al)
14
Geometrical configuration
System optimization
OBJECTIVE: To identify the limits in power for a
required resolution as a function of the magnification
of the system
METHOD
 Assess achievable spatial resolution
 Function of magnification and X-ray energy
 Spatial resolution as cutoff frequency for values MTF of
0.5 and 0.2
EFFECT ON IMAGE QUALITY
Increase of focal spot size with power has a negative impact on
resolution
15
Optimal Acquisition Protocol
Frequency at MTF 50%
f (lpmm)
Maximum Resolution @MTF20*
Frequency at MTF 20%
f (lpmm)
degradation
for Mmin = 1.56 **:
49,4 um for Pmin = 6W
54,5 um for Pmax = 50W
~ 9%
for Mmax = 3.0 **:
19,5 um for Pmin = 6W
41,9 um for Pmax = 50W
*
~ 43%
* Dexela model MTF provided by manufacturer
** FOV geometry, distances for Uc3m test bench
16
Contributions
17
Towards Spectral CT
HYPOTHESIS:
Spectral CT increases the sensitivity of X-ray
projection images in the diagnosis of pulmonary
infectious disease and allow for a more accurate
assessment of granuloma distribution
TECHNOLOGY TO DEPLOY:
Micro-CT demonstrator for the study of low dose Xray cone beam tomography based on hybrid pixels
that resolve energy levels
Which implies…
… the evolution to photon-counting detector
18
Work progress report from
UC3M
Ana Ortega-Gil
Juan José Vaquero
Supported by the INFIERI EU project n°317446, FP7-PEOPLE-2012-ITN