CA2513592A1 - Photon counting digital imaging apparatus and system - Google Patents

Abstract

The Inventors have designed a photon counting chip (PCC) with an active-matrix array of selective photon counting pixels (PCPs). The PCC provides low noise, high sensitivity, better weighing of information, wide dynamic range, high readout rate, energy selectivity, and scalability. The PCC
also offers the capability of making large area detectors with its 3-side buttable design. Each PCP is comprised of a detector, a charge-integrating amplifier (CIA), a window comparator, a decision-making unit (DMU), a mode selector, and a pseudo-random counter.

Description

PHOTON COUNTING DIGITAL IMAGING APPARATUS AND
SYSTEM
FIELD OF THE INVENTION
The invention generally relates to method and apparatus for detecting and counting photons of varying wavelengths.
BACKGROUND
Conventional film screen mammography does not detect all cancers and as many as 20%
of cancers that become clinically evident over the course of a year will not have been visible by screening mammography performed within that year [1]. A major factor contributing to this limitation is the anatomical noise created by the overlap of normal structures within the breast, which are superimposed on each other in a standard two-dimensional mammogram. Initial studies show that tomosynthesis has the ability to reveal 16% more cancers than conventional mammography and reduce false positives by 85%

[2] even at a radiation dose to the patient comparable to a conventional two-view mammogram. In order not to increase patient dose, the imaging system has to be able to readout multiple (ideally 100) images with each image using only 1% of the normal dose. However, this puts an extreme requirement on the detector in regards to the parasitic amplifier noise level. Almost all mammographic detectors currently available are based on photon integration techniques and are amplifier noise limited in some part of their operating range (i.e. dark part of image) even at normal doses, resulting in additional radiation necessary to make up for the deficiency. Increasing the number of readouts makes this problem much worse. Overcoming amplifier noise in the subdivided images is therefore the key problem in implementing a practical tomosynthesis system.
This background information is provided for the purpose of making known information believed by the applicant to be of possible relevance to the present invention. No admission is necessarily intended, nor should be construed, that any of the preceding information constitutes prior art against the present invention.
SUMMARY OF THE INVENTION
An object of the present invention is to provide a photon counting pixel architecture. In accordance with an aspect of the present invention, there is provided an alternative approach to traditional photon integration systems is photon counting where the value of each image pixel is equal to the number of photons that interact with the detector. In the photon counting method, the photon-generated charge is integrated using the charge integrating amplifier (CIA) while the output of the CIA is being processed by the window comparator. The decision-making unit (DMU) is responsible to decide whether the signal level of the photon-generated charge falls within the specified window in which case the counter will be incremented. Thus the window comparator determines the band of photon energies that will be detected and assigned the same weight (i.e.
unity) where photons with out-of band energies will be rejected. The main features of the selective photon counting pixel (PCP) are low noise, high sensitivity, better weighing of information, wide dynamic range, high readout rate, energy selectivity, and scalability.
The effect of amplifier noise is entirely independent of frame readout rate since readout is performed entirely in the digital domain after all photons have been detected.
This noise suppression compared to photon integrating detectors gives higher signal-to-noise (SNR) ratios and consequently increased device sensitivity.
The PCC provides better weighting of information from photons of different energies.
Higher energy photons deposit more charge in the detector than low energy ones so that in an energy integrating detector the higher energy photons are weighted higher. In mammography, the higher part of the 30 kVp energy spectrum, after having passed through the patient, carries less useful information than the lower energy part, but has more weight in the image formation of a photon integrating system. In a photon counter, all events are weighed equally.
Due to predetermined thresholds used to form the window, photon-counting systems are inherently linear and can offer very large dynamic ranges. This becomes important in mammography where the high dynamic range associated with the modality is challenging to achieve even in state-of the-art amorphous silicon (a-Si) flat panel imagers [3][4], usually because of pixel saturation at higher X-ray inputs. In contrast, the dynamic range of photon counting devices can be much higher (at the low count rate there is no limit and at high count rate by the separation of pulses and capacity of counters).
The PCC has two modes of operation: 1 ) detection mode, and 2) readout mode.
In the detection mode the number of the photons counted by each PCP is stored in the pseudo-random counter. In the readout mode, all PCPs in the PCC will be connected to each other and an external clock is generated to shift counter values serially out.
Again note that readout is performed entirely in the digital domain; hence, fast readout rate.
Each PCP is equipped with a window comparator which allows the capability of counting photons with a specified energy range. In mammography we are interested in low-energy x-ray photons since low x-ray energies provide the best differential attenuation between tissues meaning that the x-ray attenuation differences between the normal tissue and the cancerous lesion is highest at very low x-ray energies (10 to 30 keV). The PCC
can be configured to count photons with the user-specified energy range.
Each PCP is equipped with all the intelligence needed to perform photon counting and operates independently of any other PCP; hence, increasing the number of PCPs in a PCC
is a simple task and will not affect design complexity nor will it degrade overall performance.
BRIEF DESCRIPTION OF THE FIGURES

The invention will now be described with references to the drawings in which:
Figure 1 illustrates the block diagram of the PCP.
Figure 2 shows the schematic of the PCP.
Figure 3 shows the signal waveforms for the CIA, the window comparator, and the DMU.
Figure 4 illustrates the four scenarios that need to be considered when calculating the required DMU clock period or the delay time tdelay~
Figure 5 depicts a 3x3 active-matrix array of a 3-side buttable PCC.
Figure 6 shows a detector configuration with a 2x3 array of three-side buttable PCCs.
DETAILED DESCRIPTION OF THE INVENTION
The photon counting chip (PCC) is an active-matrix array of photon counting pixels (PCPs) where each PCP is interfaced to a detector. The detector is responsible to deposit electric charge when interacted with a photon. For photon counting systems, the detector is required to have carrier multiplication and the number of carriers generated must be greater than that of input noise.
PCP OPERATION
As illustrated in Figure l, each PCP is comprised of a detector, a charge-integrating amplifier (CIA), a window comparator, a decision-making unit (DMU), a mode selector, and a pseudo-random counter. The circuit schematic is shown in Figure 2.
Each photon counting pixel (PCP) is connected to a detector where the detector is responsible to deposit electric charge when interacted with a photon. For photon counting systems, the detector is required to have carrier multiplication and the number of carriers generated per photon must be greater than that of input noise.
Each PCP has two distinct modes: 1 ) detection mode, and 2) readout mode. In the operating mode, which is selected by the mode selector, the photon-generated carriers in the pixel are integrated by the CIA while the output of the CIA is being compared to three thresholds, Vth-hi,Vth-lo, and Vref where Y",-,,; >_ V~,,-,o >_ V,.ef > peak output noise voltage of the CIA
Thresholds Vth-hi and Vth_,o are used to form the desirable window and remain unchanged for the duration of the detection period. After comparisons are performed, outputs of the window comparator are passed to the DMU. The DMU is responsible to decide whether the counter is to be incremented and also whether CIA's feedback capacitor Cf is required to be discharged. Since it takes time td to integrated all photon-generated carriers on the feedback capacitor Cf, the DMU has to delay its decision making process by tdelay. Note that taelay is also the period for the DMU clock. The operation of the PCP in detection mode can be easily explained using the following pseudo code:
while ( operating in detection mode ) //Integrate carriers lf( Vref C VCIA out C Vth hi) wait for tdelay; //Too early to decide.
1f( Vref ~ VCIA out ~ Vth_to ) discharge feedback capacitor Cf;
continue;
else lf( V~h to C VCIA out C Vth hi f increment pseudo-random counter;
discharge feedback capacitor Cf;
continue;
else if ( V~,A ~~t > Vth hi ) f discharge feedback capacitor Cf;
else //Wait for a photon to arrive.

Figure 3 demonstrates the operation of the PCP by means of waveforms for three cases:
1 ) Vc,a out exceeds the upper window threshold Vin h;, 2) Vc~,a oUt falls within the window, and 3) Vc,n o~, just exceeds the reference voltage Vref during tae,a,,. For the first and third case, the DMU has decided not to increment the counter but to reset the feedback capacitor Cf. However for the second case, the DMU has decided to both increment the counter and discharge the feedback capacitor.
DMU DELAY TIME AND PCP PHOTON COUNT RATE
The delay time tdelay, which as mentioned before is also the period for the DMU clock, is proportional to td which is the time required for all the photon-generated carriers to travel detector thickness and appear at the non-inverting terminal of the CIA.
To theoretically calculate tdelay of the PCP we need to consider four scenarios as shown in Figure 4. The time t",;n in Figure 4 is the time during which the output of the CIA falls within the window. It be calculated by knowing the number of Garners Qd generated during the time interval td by the detector, the value of the feedback capacitor Cf, and the characteristic of the window comparator C~ OV t~
tWlll For the first scenario of Figure 4, the minimum DMU clock period (or taelay) must be equal to tW;n since the decision making process has to be delayed until Vc,A out exceeds V,h n.
(DMU clk period~s~, _ (tdela ~ ~ tWlll y sa Note from the waveforms shown in Figure 3 that it takes three clock cycles for the DMU
to perform its decision making process after V~,A o", exceeds V,h ,o. Thus, the photon count rate is inversely proportional to the time it takes to reach from VCIn_bias to Vtn o, also known as too, plus three DMU clock cycles (photon count rate~s~, S
too + 3t",;" photons per second].
For the second scenario of Figure 4, the minimum DMU clock period can no longer be equal to t~";" since after the feedback capacitor is reset, there still exist some photon-generated carriers that are not yet integrated. The period of the DMU clock in this case must be large enough to ensure that all the photon-generated carriers have reached the non-inverting terminal of the CIA at the end of the third DMU clock cycle (DMU clk period~s~2 >_ ~~'A ont-naz - ytn ro x t 3" .
Thus, the photon count rate for the second scenario is (photon count rate~s~ z <_ I ' vCIA ottt maz ytle to too + - d x twt"
As for the third scenario, note from Figure 3 waveforms that it takes two clock cycles for the DMU to perform its decision making process after V~~ o"t exceeds Vt,, a.
Therefore the minimum DMU clock period should be half of the time tw;n ~DMU clk period~s~, >_ t'"~"

The photon count rate for the third scenario is (photon count rate~s~3 <_ tLO '~ twr»
Finally, for the fourth scenario, the minimum DMU clock period must be equal to too (DMU clk period~s~a >_ tLo Thus, the photon count rate for the fourth scenario is photon count rate~s~a <_ 1 4t~o PCC DESIGN
After all the photons are counted, the mode selector switches to readout mode where all the PCPs in the PCC are connected to each other to serially shift counter values out. A
microcontroller is used to capture the serially shifted bits and to form a matrix with each element in the matrix representing the counter value of the corresponding PCP.
Figure 5 depicts a 3x3 active-matrix array of a 3-side buttable PCC. The design is scalable to achieve a large l8cm x 24 cm tiled imager. Figure 6 shows a detector configuration with a 2x3 array of three-side buttable PCPs.

The embodiments of the invention being thus described, it will be obvious that the same may be varied in many ways. Such variations are not to be regarded as a departure from the spirit or scope of the invention, and all such modifications as would be obvious to one skilled in the art are intended to be included within the scope of the following claims.
OTHER REFERENCES
[ 1 ] R. Freifelder and J. S. Karp, Dedicated PET Scanners for Breast Imaging, Phys. Med.
Biol. 42, 2463-2480 (1997).
[2] E. Rafferty, "Tomosynthesis: New Weapon in Breast Cancer Fight," Guest editorial in Decisions in Imaging Economics, The Journal of Imaging Technology Management, April 2004.

[3] K.S. Karim, A. Nathan, J.A. Rowlands, "Amorphous silicon pixel amplifier with VT
compensation for low noise digital fluoroscopy," in IEEE International Electron Devices Meeting (IEDM) Technical Digest, 215-218 (2002).

[4] J.A. Rowlands, J. Yorkston, "Flat panel x-ray imagers for digital radiography", Chapter 4 Medical Imaging Volume 1, Physics and Psychophysics, SPIE, Bellingham, (2000)

Claims (2)

1. A digital imaging apparatus comprising:
a) a detector for generating a first signal in response to photons incident thereupon; and b) photon counting readout circuitry coupled to said detector for receiving said first signal and for generating a second signal representative of said first signal.

2. A digital imaging system comprising:
a) multiple copies of photon counting digital imaging apparatus arranged to create a larger area imager