Universitt ErlangenNrnberg Physikalisches Institut Energy dependent imaging properties
Universität Erlangen-Nürnberg Physikalisches Institut Energy dependent imaging properties of the Medipix 2 X-ray-detector Vertex 2007 Lake Placid September 28, 2007 Thilo Michel Institute of Physics University of Erlangen-Nuremberg, Germany
Universität Erlangen-Nürnberg Outline The Medipix 2 detector • Medipix 2: a detector for x-ray imaging • The working principle • Processes in the sensor layer Physikalisches Institut Imaging properties 2 • Spatial resolution Using the energy resolution • Reconstruction of incident x-ray spectra • Detective-Quantum-Efficiency • Material decomposition • Energy response function • Outlook: Medipix 3
Physikalisches Institut Universität Erlangen-Nürnberg The Medipix 2 -Detector: a direct-converting, hybrid photon counting pixel detector 3 E • ASIC/Sensor: – Development: International Collaboration with seat at CERN – Bump-bonded with Pb/ Sn – 65536 pixels in 256 columns and 256 rows – Pixel pitch: 55 µm – Size of the matrix: 14 mm – 0. 25 µm CMOS • Sensor: – Materials: Si, Ga. As, Cd. Te – Bias voltage: 150 V (300 µm Si) – 2 x 2 -version 14 mm
Physikalisches Institut Universität Erlangen-Nürnberg Medipix 2: a photon counting ASIC 4 Pixel electronics Working principle • Idea: –Counting of each photon with energy deposition above threshold or in energy window –Counting is noiseless 14 bits • Analog part: –Charge sensitive pre-amplifier – 2 discriminators per pixel (window mode, single threshold mode) • Digital part: – 1 counter –Max. 800 million photons per image –Pseudo-random counters act as shift registers during readout –Read out times: 9 msec (serial), approx. 265 µsec (parallel)
Universität Erlangen-Nürnberg Outline The Medipix 2 detector • Medipix 2: a detector for x-ray imaging • The working principle • Detecting X-ray photons Physikalisches Institut Imaging properties 5 • Spatial resolution Using the energy resolution • Reconstruction of incident x-ray spectra • Detective-Quantum-Efficiency • Material decomposition • Energy response function • Outlook: Medipix 3
Edge spread function Line spread function (lsf) Intensity [a. u. ] Universität Erlangen-Nürnberg Measurement of the Modulation-Transfer-Function (MTF) with the edge method x[mm] MTF (90 k. Vp, E thr=6 ke. V) Physikalisches Institut Ideal MTF (sinc) 6 Measurement Simulation
Physikalisches Institut 7 MTF for low spatial frequencies Schematic of the detector assembly Simulation: without glue Sim. : 20 µm Ag MTF Universität Erlangen-Nürnberg The Low-Frequency-Drop in the MTF Heat conducting glue Sim. : 30 µm Ag Measurement PCB Spatial frequency f [lp/mm] • Production of fluorescence photons in the silver containing glue between ASIC and PCB • Triggering of pixels sometimes far away from the production vertex of the fluorescence photons • This long range effect leads to a reduction of MTF at low spatial frequencies
8 Pixel profile [counts] Physikalisches Institut Universität Erlangen-Nürnberg Simulation: The pixel profile (response to line source) in dependence on the discriminator threshold 6 ke. V threshold Fluorescences from bump bond 15 ke. V threshold Fluorescences from neighbouring bump bond 45 ke. V threshold Distance from pixel center [mm] Tube voltage: 90 k. Vp, tungsten anode
Physikalisches Institut 9 Measurements Simulations 45 ke. V threshold MTF Universität Erlangen-Nürnberg The dependency of the MTF on the discriminator threshold: higher spatial resolution at higher discriminator thresholds 15 ke. V threshold 6 ke. V threshold Spatial frequency [lp/mm] Excellent agreement of simulation and measurement Tube voltage: 90 k. Vp, tungsten anode
Universität Erlangen-Nürnberg Outline The Medipix 2 detector • Medipix 2: a detector for x-ray imaging • The working principle • Detecting X-ray photons Physikalisches Institut Imaging properties 10 • Spatial resolution Using the energy resolution • Reconstruction of incident x-ray spectra • Detective-Quantum-Efficiency • Material decomposition • Energy response function • Outlook: Medipix 3
Physikalisches Institut Universität Erlangen-Nürnberg A model to describe the Detective-Quantum-Efficiency (DQE) of a photon counting pixel detector 11 Problem: Multiple counting affects the image noise The (average) multiplicity • Number of counts in one pixel varies according to Poisson-statistics • Charge-Sharing: Numbers of counts in different pixels are not independent random variables • Image noise is depending on fraction of multiple counts • Therefore: Influence on the DQE T. Michel et al. , „A fundamental method to determine the signal-to-noise ratio (SNR) and detective quantum efficiency (DQE) for a photon counting pixel detector“, Nuclear Instruments and Methods A, Volume 568, Issue 2, p. 799 -802 (2006)
Physikalisches Institut Universität Erlangen-Nürnberg Simulations of the average multiplicity and the DQE(0) 12 40 k. V spectrum, Mo-Anode 100 ke. V <m>2 / <m 2> e DQE(0) <m 2> <m>
Universität Erlangen-Nürnberg Outline The Medipix 2 detector • Medipix 2: a detector for x-ray imaging • The working principle • Detecting X-ray photons Physikalisches Institut Imaging properties 13 • Spatial resolution Using the energy resolution • Reconstruction of incident x-ray spectra • Detective-Quantum-Efficiency • Material decomposition • Energy response function • Outlook: Medipix 3
Physikalisches Institut 14 Energy response at 59. 3 ke. V (tungsten Ka 1) Measurement Simulation Counts Universität Erlangen-Nürnberg Simulations and measurements of the energy response Sensor only Energy deposition [ke. V] • Response function R depends on various parameters: R = R(E‘, Eγ, Vbias, d. Sensor, p, sensormaterial) • But: Can be described very well with our simulations • Validation of simulation code in measurements
Universität Erlangen-Nürnberg Outline The Medipix 2 detector • Medipix 2: a detector for x-ray imaging • The working principle • Detecting X-ray photons Physikalisches Institut Imaging properties 15 • Spatial resolution Using the energy resolution • Reconstruction of incident x-ray spectra • Detective-Quantum-Efficiency • Material decomposition • Energy response function • Outlook: Medipix 3
Physikalisches Institut 16 Response spectrum of polychromatic irradiation Number of counts or photons Universität Erlangen-Nürnberg Methods for the reconstruction of incident x-ray spectra with high flux Medipix 2 response spectrum Spectrum-Stripping-Method Subtract the monoenergetic response functions successively: … Impinging spectrum Energy [ke. V] R(Ei|, Ej) : the probability that a photon of energy Ej causes an energy deposition of Ei| Matrix-Inversion-Method Best estimate for impinging spectrum is given by:
17 Medipix 2 -response spectrum Impinging spectrum 80 k. Vp with spectrum stripping Number of photons [a. u. ] Number of counts or photons Measured energy deposition spectrum Energy [ke. V] Theoretical Reconstructed Energy[ke. V] Simulated response functions 120 k. Vp with matrix inversion Number of photons [a. u. ] Physikalisches Institut Universität Erlangen-Nürnberg Results of the spectrum reconstruction methods Theoretical Reconstructed Energy [ke. V]
Universität Erlangen-Nürnberg Outline The Medipix 2 detector • Medipix 2: a detector for x-ray imaging • The working principle • Detecting X-ray photons Physikalisches Institut Imaging properties 18 • Spatial resolution Using the energy resolution • Reconstruction of incident x-ray spectra • Detective-Quantum-Efficiency • Material decomposition • Energy response function • Outlook: Medipix 3
Physikalisches Institut Universität Erlangen-Nürnberg Application of reconstruction methods: Quantitative determination of the material composition of irradiated objects 19 Transmission for different incident x-ray energiesi E Mass attenuation coeffizient Application of the pseudoinverse matrix M gives best estimate of areal densities p of the involved materials Iteration Areal density of material j The materials-matrix Energy Material
Physikalisches Institut Universität Erlangen-Nürnberg First experimental try of material reconstruction with iodine, plastics (PMMA) and alumunium 20 • • 50 k. Vp, tungsten anode Threshold scan 450 m. As/image charge in tube 200. 000 counts per pixel at lowest threshold in directly irradiated ROI The phantom: view from tube PMMA Projective photon counting image
Physikalisches Institut Universität Erlangen-Nürnberg Material reconstructedimages for each base material obtained with simulated energy response functions 21 Expected images Iodine PMMA Aluminium Material reconstructed images [g/cm 2] Iodine [g/cm 2] PMMA Reconstruction: energy deposition range: 20 -50 ke. V; energy bin size: 2 ke. V Aluminium
Universität Erlangen-Nürnberg Outline The Medipix 2 detector • Medipix 2: a detector for x-ray imaging • The working principle • Detecting X-ray photons Physikalisches Institut Imaging properties 22 • Spatial resolution Using the energy resolution • Reconstruction of incident x-ray spectra • Detective-Quantum-Efficiency • Material decomposition • Energy response function • Outlook: Medipix 3
Physikalisches Institut Universität Erlangen-Nürnberg Medipix 3 will have a high potential for applications of the material reconstruction method 23 Charge-Summing • Real-time communication in each 2 x 2 neighbour-hood • Detection of the total released charge in one node ( ) Simulated energy response for 15 ke. V Charge-Summing. Mode Single-Pixel-Mode Combined with true color mode (110 µm pixel pitch in the sensor): this spectroscopic detector is very good for material reconstruction
Physikalisches Institut Universität Erlangen-Nürnberg High resolution images taken in collaboration with the Univeristy of Gent will be shown in the talk of Bert Masschaele 24
Physikalisches Institut Universität Erlangen-Nürnberg Summary 25 ● ● ● Simulations predict behaviour of Medipix 2 very well Image quality – Spatial resolution (MTF) is reduced by fluorescence photons, diffusion and range of electrons – Image noise is correlated due to multiple counting which reduces DQE Spectral information – Incident x-ray spectrum at high fluxes can be measured – Selective images of materials can be calculated from images taken at different thresholds – First experiments of material reconstruction have been successful – Medipix 3: high potential for applications of the material reconstruction
Physikalisches Institut Universität Erlangen-Nürnberg Backup 26
Physikalisches Institut Universität Erlangen-Nürnberg Advantages and disadvantages of the Medipix 2 in x-ray imaging applications 27 Advantages Disadvantages • Noise free dark image • Small size • No read out noise • • No blooming Tiling to large arrays without dead zones or large edge pixels not yet done • No after glow • Charge-Sharing • Exact linearity at low rates • • High spatial resolution Multiple counting: reduction of DQE • Rate limitation at very high rates • High framerate • Energy sensitive
Physikalisches Institut Universität Erlangen-Nürnberg The reacting photon deposits 100 ke. V in four neighbouring pixels 28 100 ke. V
Physikalisches Institut Universität Erlangen-Nürnberg Each charge sensitive preamplifier generates a current signal that is proportional to the collected charge 29 10 ke. V 30 ke. V 20 ke. V 50 ke. V
Physikalisches Institut Universität Erlangen-Nürnberg 4 current signal copies are generated and sent to the corners (nodes) 30 10 ke. V 30 ke. V 20 ke. V 50 ke. V
Physikalisches Institut Universität Erlangen-Nürnberg The 4 current signals reach the nodes and there … 31 10 10 30 30 10 20 10 30 20 50 30 50 20 20 50 50
Physikalisches Institut Universität Erlangen-Nürnberg … they are summed 32 10 40 30 30 100 80 20 70 50
Physikalisches Institut Universität Erlangen-Nürnberg The vertical arbiters point to the vertical node with the higher sum signal 33 10 40 30 30 100 80 20 70 50
Physikalisches Institut Universität Erlangen-Nürnberg The horizontal arbiter compare the maximum on one side with the maximum on the other side 34 10 40 30 30 100 80 20 70 50
Physikalisches Institut Universität Erlangen-Nürnberg There is only one node with all the surrounding arbiters pointing to it and … 35 10 40 30 30 100 80 20 70 50
Physikalisches Institut Universität Erlangen-Nürnberg … this node counts the photon if the threshold is exceeded 36 100
37 Reconstructed spectra with both methods Number of photons [a. u. ] Physikalisches Institut Universität Erlangen-Nürnberg Differences occur if the bin size for reconstruction comes closer to the noise level of the detector Energy [ke. V]
Physikalisches Institut Universität Erlangen-Nürnberg Measurement of the average multiplicity 38 Measurement Distribution of event multiplicities • Monochromator setup for 59 ke. V, 8 ke. V threshold, 700 µm Si sensor • Framerate of 20 Hz • Cluster analysis determines multiplicity of each event Comparison: simulation and experiment
Physikalisches Institut Universität Erlangen-Nürnberg Material reconstruction in simulated CT with an ideal photon counting detector and „friendly“ materials (K-edges) 39 I Gd H 2 0 Iodine Gadolinium Oxygen Hydrogen Iodine Water Gadolinium
Physikalisches Institut 40 Ideal detector: sinc(55µm) Measurements Simulations MTF Universität Erlangen-Nürnberg The MTF also depends on the impinging spectrum: higher spatial resolution at lower energies 90 k. Vp 40 k. Vp Spatial frequency [lp/mm]
Physikalisches Institut 41 20 ke. V, no fluorescences Ideal detector: sinc(55 µm) 20 ke. V threshold MTF Universität Erlangen-Nürnberg Fluorescence photons and charge sharing have strong impact on the MTF 6 ke. V threshold Suppression of the influence of fluorescences Suppression of charge sharing (diffusion, cont. energy loss) Spatial frequency [lp/mm] Tube voltage: 90 k. Vp, tungsten anode
Physikalisches Institut Universität Erlangen-Nürnberg The modulation transfer function (MTF): a key performance indicator of an imaging detector 42 Spatial domain Spatial frequency domain Modulation-Transfer. Function (MTF)
43 Cd. Te Linear attenuation coefficient [1/mm] Physikalisches Institut Universität Erlangen-Nürnberg First interaction of an x-ray photon with sensor Si: total Si: Rayleigh Ga. As Si: Photoeffect Si: Compton Photon energy [ke. V]
- Slides: 43