Transcript LaureaSpecialistica - INFN

A silicon microstrip system with
the RX64DTH ASIC for dual
energy radiology
• Introduction
– Why digital?
– Why dual energy?
• Experimental setup
• Image acquisition
• Image processing and results
The Collaboration
1) University of Eastern Piedmont and INFN, Alessandria, Italy
L. Ramello;
2) University and INFN, Torino, Italy
P. Giubellino, A. Marzari-Chiesa, F. Prino;
3) University and INFN, Ferrara, Italy;
M. Gambaccini, A. Taibi, A. Tuffanelli, A. Sarnelli;
4) University and INFN, Bologna, Italy
G. Baldazzi, D. Bollini;
5) AGH Univ. of Science and Technology, Cracow, Poland
W. Dabrowski, P. Grybos, K. Swientek, P. Wiacek;
6) University of Antwerp, Antwerp, Belgium
P. Van Espen;
7) Univ. de los Andes, Colombia
C. Avila, J. Lopez Gaitan, J.C. Sanabria;
8) CEADEN, Havana, Cuba
A.E. Cabal, C. Ceballos, A. Diaz Garcia;
9) CINVESTAV, Mexico City, Mexico
L.M. Montano;
Introduction: why digital ?
• Digital radiography has well known advantages over
conventional screen-film systems
– Enhance detecting efficiency w.r.t. screen-film
– Image analysis
– Easy data transfer
Introduction: why dual energy ?
• Dual energy techniques
Based on different energy
dependence of the
absorption coefficient of
different materials
• GOAL: improve image contrast
Enhance detail
visibility (SNR)
Decrease dose
to the patient
Decrease contrast
media concentration
Example 1: dual energy
mammography
Example 1: dual energy
mammography
E  15-20 keV:
Signal from cancer tissue
deteriorated by the
adipose tissue signal
E  30-40 keV
Cancer tissue not visible,
image allows to map glandular
and adipose tissues
Example 2: angiography
•Angiography = X-ray examination of blood vessels
 determine if the vessels are diseased, narrowed or blocked
 Injection of a contrast medium (Iodine) which absorbs X-ray
differently from surrounding tissues
•Coronary angiography
 Iodine must be injected into the heart or very close to it
 A catheter is inserted into the femoral artery and managed up to the
heart
→Long fluoroscopy exposure time to guide the catheter
→Invasive examination
•Why not to inject iodine in a peripheral vein?
 Because lower iodine concentration would be obtained, requiring
longer exposures and larger doses to obtain a good image
 But, if the image contrast could be enhanced in some way…
Example 2: angiography at the
iodine K-edge (II)
Iodine injected in
patient vessels acts as
radio-opaque contrast
medium
Dramatic change of
iodine absorption coeff. at
K-edge energy (33 keV)
Subtraction of 2 images taken with photons of
2 energies (below and above the K-edge)
→ in the resulting image only the iodine signal
remains and all other materials are canceled
Experimental setup
• To implement dual energy imaging we need:
• a dichromatic beam
• a position- and energy-sensitive detector
Quasi-monochromatic beams
• ordinary X-ray tube + mosaic crystals
• instead of truly monochromatic
synchrotron radiation
Advantages: cost, dimensions, Linear array of silicon microstrips +
availability in hospitals
electonics for single photon counting
•Binary readout
• 1 or 2 discriminators (and
counters) per channel
• Integrated counts for each pixel
are readout
• Scanning required to build 2D image
Experimental setup: beam
Two spatially
separated
beams with
different
energies
 E-DE and
E+DE obtained
in 2 separate
beams
Bragg Diffraction on Highly
Oriented Pyrolitic Grafite Crystal
n.h.c
E 
B 2d sin 
B
Double slit
collimator
W anode tube
Experimental seup: Single Photon
Counting System
X-rays
current pulses
N. I. I/O cards PCI-DIO96
and DAQCard-DIO-24
100 m
data,
control
Silicon strip detector
Integrated
circuit
PC
• Fully parallel signal processing for all channels
• Binary architecture for readout electronics
1 bit information (yes/no) is extracted from each strip
Threshold scans needed to extract analog information
• Counts integrated over the measurement period transmitted to DAQ
Detecting system
Silicon microstrip detector
each strip is an independent detector
which gives an electric signal when an Xray photon crosses it and interacts with
a silicon atom
4 cm
Chip RX64 → counts incident photons on
each strip of the detector
6.4 mm
10 strip = 1 mm
micro-bondings
Knowing from which strip the electric
signal comes from,the position of the
incoming X-ray phonton is reconstructed.
Why silicon detectors?
Main characteristics of silicon detectors:
• speed of the order of 10 ns
• spatial resolution of the order of 10 m
•small amount of material
 0.003 X0 for a typical 300 m thickness
• excellent mechanical properties
• good resolution in the deposited energy
 3.6 eV of deposited energy needed to create
a pair of charges, vs. 30 eV in a gas detector
Silicon sensor diode
•The impinging ionizing particles generate electron-hole pairs
•The impinging photons which interact in the detector volume create an
electron (via Photoelectric, Compton or Pair Production)
•The electron ionizes the surrounding atoms generating electron-hole pairs
• Electron and holes drift to the electrodes under the effect of the
electric field present in the detector volume.
•The electron-hole current in the detector induces a signal at the
electrodes on the detector faces.
Metal contact
photon
Charged particle
P+-type implant
n-type bulk
electron
photoelectron
hole
n+-type implant
E
-V
Reverse
bias
+V
Why reverse biased diode?
•The amount of charge deposited in the silicon detector is very small
≈5500 electrons are produced by a 20 keV photons making photoelctric
effect in the silicon
Forward-biased junction: the signal would be masked by the
fluctuations of the current which the applied field makes flow even
in high resistivity, hyper-pure silicon.
NOT GOOD
Reverse-biased junction: allows to obtain the necessary electric
field and only a very small “dark” current also at room temperature.
junction
-V
depleted region
+V
Increasing the
polarization voltage,
it is possible to
extend the depletion
layer down to the
backplane.
To have full efficiency,
the polarization voltage
must be high enough to
deplete the full
detector thickness
(typically 300 m)
Silicon Microstrips detectors
• A micro-strip detector is a silicon detector segmented in long, narrow
elements.
•Each strip is an independent p-n reverse-biased junction
• Provides the measurement of one coordinate of the particle’s
crossing point with high precision (down to 1 m).
DC coupling to electronics
Al
SiO2
AC coupling to electronics
Al
N-type substrate
P+
n+
P+
SiO2
Experimental setup: RX64 chip
Cracow U.M.M. design - (28006500 m2) - CMOS 0.8 µm process
(1) 64 front-end channels
a) preamplifier
b) shaper
c) 1 or 2 discriminators
(2) (1 or 2)x64 pseudo-random counters (20-bit)
(3) internal DACs: 8-bit threshold setting and 5-bit for bias settings
(4) internal calibration circuit (square wave 1mV-30 mV)
(5) control logic and I/O circuit (interface to external bus)
2
Detector
1
5
4
3
System calibration setup in Alessandria
Detector in Front config.
Fluorescence target
(Cu, Ge, Mo, Nb, Zr, Ag, Sn)
Cu anode X-ray tube
→ X-ray energies = characteristic lines of target material
System calibration
Mo K
Counts
150
Sn K
Ge K
100
Ag K
Cu K
Mo K
Ag K
Rb K
50
0
100
200
Source Am+Rb target
Source Am+Mo target
Source Am+Ag target
Tube+Cu target
Tube+Ge target
Tube+Mo target
Tube+Ag target
Tube+Sn target
Sn K
300
400
500
Threshold (mV)
241Am
source with rotary target holder (targets: Cu, Rb, Mo, Ag, Ba)
Cu-anode X-ray tube with fluorescence targets (Cu, Ge, Mo, Ag, Sn)
System
Tp
GAIN
V/el.
ENC
Energy
resolution
6 x RX64
0.7 s
64
≈170 el.
≈0.61 keV
6 x RX64DTH
0.8 s
47
≈ 200 el.
≈0.72 keV
Imaging test
1-dimensional array of strips → 2D image obtained by scanning
Test Object
5 mm
Collimator (0.5 mm)
Detector
Cd-109 source (22.24 keV)
Imaging test
Pasos
Scanning
1-dimensional array of strips → 2D image obtained by scanning
210
200
190
180
170
160
150
140
130
120
110
100
90
80
70
60
50
24,00
21,00
18,00
15,00
12,00
9,000
6,000
3,000
0
0
10
20
30
40
Canales
50
60
Dual energy imaging
• K-edge subtraction imaging with contrast medium
 Cancel background structures by subtracting 2 images taken at energies just
below and above the K-edge of the contrast medium
 Suited for angiography at iodine (gadolinium) K-edge
- Cancel background structures to enhance vessel visibility
 Possible application in mammography (study vascularization extent)
- Hypervascularity characterizes most malignant formations
• Dual energy projection algorythm
 Make the contrast between 2 chosen materials vanish by measuring the
logarithmic transmission of the incident beam at two energies and using a
projection algorithm [Lehmann et al., Med. Phys. 8 (1981) 659]
 Suited for dual energy mammography
– remove contrast between the two normal tissues (glandular and
adipose), enhancing the contrast of the pathology
– Single exposure dual-energy mammography reduces radiation
dose and motion artifacts
Angiography setup
X-ray tube with dual energy
output
Phantom
Detector box with 2
collimators
1.
X-ray tube + mosaic crystal and 2 collimators to provide dual-energy output
- E1= 31.5 keV, E2 =35.5 keV (above and below iodine k-edge)
2.
Detector box with two detectors aligned with two collimators
3.
Step wedge phantom made of PMMA + Al with 4 iodine solution filled
cavities of 1 or 2 mm diameter
pixels
15
10
5
6
15
5
pixels
3
Conteggi (x10 )
Angiographic test results (I)
4
3
10
2
5
1
0
0
200
pixels
100
0
0
300
E = 31.5 keV
100
E = 35.5 keV
1,0
Counts / Max.Counts
0,8
0,6
logarithmic subtraction
0,4
C1  ln N35.5   C2  ln N31.5 
0,2
Measurement
Simulation
0,0
0
50
100
150
200
250
300
Counts / Max.Counts
Conc(I) = 370 mg/ml
E = 31.5 KeV
1,0
Conc(I) = 370 mg/ml
E = 35.5 KeV
0,6
0,4
0,2
Measurement
Simulation
0,0
0
50
100
350
-0.4
-0.6
150
200
250
Strip Number
15
10
5
-0.8
0
0
100
200
pixels
300
ln[count(E=35.5Kev)] - ln[count(E=31.5Kev)]
-0.2
pixels
log conteggi
0.0
Measurement
Simulation
Conc(I) = 370 mg/ml
0,2
0,0
-0,2
-0,4
-0,6
-0,8
-1,0
0
50
100
150
200
Strip Number
300
0,8
Strip Number
Phantom
structure not
visible in
final image
200
pixels
250
300
350
300
350
Angiographic test results (II)
5
-0.8
0.0
-0.2
10
5
-0.3
0
0.15
0.10
0.05
0.00
-0.05
-0.10
-0.15
0
0
100
200
300
100
200
300
5
0
100
200
300
pixels
Conc = 92.5 mg / ml
cavità 4 teor.
cavità 3 teor.
cavità 2 teor.
cavità 1 teor.
80
SNR
100
pixels
Conc = 370 mg / ml
10
0
0
pixels
15
pixels
0.1
-0.1
log conteggi
-0.6
10
15
pixels
-0.4
15
pixels
-0.2
log conteggi
0.2
0.0
Conc = 23.1 mg / ml
cavità 4
cavità 3
cavità 2
cavità 1
60
40
20
0
0
100
200
Concentrazione (mg/ml)
300
400
Possible decrease of iodine concentration keeping the same rad. dose
Results with a second phantom
Phantom
0
100
200
300
140
120
0
100
200
300
140
140
300 um pixel
100
120
300 um pixel
100
120
80
100
60
80
80
60
40 60
40
20 40
20
20
0
0
100
200
300
140
0
140
0
0
120
100
300 um pixel
Digital Subtraction
Angiography
Dual Energy
Angiography
Iodine conc. = 95 mg/ml
smaller cavity
(=0.4 mm)
visible in DEA
and not in DSA
100
200
100 um pixel
300
120
100
80
80
60
60
40
40
20
20
0
0
100
200
100 um pixel
300
Conclusion and Outlook
• We have developed a single photon counting silicon detector equipped
with the RX64DTH ASIC, with two selectable energy windows
• The energy resolution of 0.8 keV (rms) is well adapted for dual energy
mammography and angiography
• We have performed mammography imaging tests with a three-material
phantom
– We have demonstrated the feasibility of contrast cancellation between two materials,
enhancing the visibility of small features in the third one
• We have performed angiography imaging tests with 2 different
phantoms and iodine contrast medium
– We have demonstrated the feasibility of logarithmic subtraction between two images,
enhancing contrast of small vessels also with lower iodate solution concentrations
• OUTLOOK:
– Increase photon statistics at high energy, optimize exposure conditions
– New detector materials, CZT?
– Tests with a more realistic mammographic phantom
• LAVORO INTERESSANTE E STIMOLANTE PER UNA TESI DI
LAUREA SPECIALISTICA
Tesi di laurea specialistica in fisica medica
per studenti delle Università di Torino e del Piemonte Orientale
Sviluppo di un rivelatore al silicio per immagini
radiologiche digitali a doppia energia
Immagine
radiografica
210
Oggetto di test
Pasos
24,00
Rivelatore a microstrip di silicio
+
Elettronica per conteggio di fotoni
200
190
180
170
160
150
140
130
120
110
100
90
80
70
60
50
21,00
18,00
15,00
12,00
9,000
6,000
3,000
0
0
10
20
30
40
50
60
Canales
Per informazioni:
Alberta Marzari
Francesco Prino
Luciano Ramello
Piano di lavoro:
• 6 mesi* al CINVESTAV, Città del Messico
• Conclusione e stesura a Torino/Alessandria
* indicativamente da metà giugno a metà dicembre ‘05

011 6707369, 0131 360154
 [email protected], [email protected], [email protected]