Medipix Family Research Articles

From the Higgs–Boson to Molecular Radiology

This paper delves into the history of MARS photon-counting CT and its technology origins in particle physics at the European Organization for Nuclear Research (CERN). The story begins at CERN, a world-class facility for research into fundamental physics. In the 1960s, charged particle experiments at CERN involved the slow and labor-intensive work of examining millions of photographs from bubble or spark chambers. In the 1970s, semiconductor detector technology changed nuclear and particle physics, and the world at large. As its name implies, semiconductor material conducts current, but this current can be modified, especially by a quantum of light or other radiation. During the development of semiconductor detector technology, the potential benefit of detecting X-ray photons became evident, leading to the development of the Medipix family of imaging devices in the 1990s. In the early 2000s, researchers from New Zealand took the Medipix detector and began to research and develop spectral molecular imaging for the clinic. Their work produced the MARS photon-counting CT, which promises to significantly advance current CT, develop a new standard of care, and improve health outcomes for millions of patients world-wide. Taking MARS to the clinic will be discussed, including investigation of potential clinical applications and the future direction of MARS technology.

X-ray imaging at synchrotron research facilities

At synchrotron facilities, many X-ray imaging and diffraction experiments require pixel detectors with minimal noise, high speed and reasonably small pixel size, all of which can be achieved with the Medipix ASIC family. So, ESRF, Diamond Light Source and DESY have developed detector systems based on Medipix. In this paper, we report on these developments, with an emphasis on the challenges involved building readout systems and the potential of the Medipix family ASICs in this field of research.

Sub-pixel electron detection using a convolutional neural network

Modern direct electron detectors (DEDs) provided a giant leap in the use of cryogenic electron microscopy (cryo-EM) to study the structures of macromolecules and complexes thereof. However, the currently available commercial DEDs, all based on the monolithic active pixel sensor, still require relative long exposure times and their best results have only been obtained at 300 keV. There is a need for pixelated electron counting detectors that can be operated at a broader range of energies, at higher throughput and higher dynamic range. Hybrid Pixel Detectors (HPDs) of the Medipix family were reported to be unsuitable for cryo-EM at energies above 80 keV as those electrons would affect too many pixels. Here we show that the Timepix3, part of the Medipix family, can be used for cryo-EM applications at higher energies. We tested Timepix3 detectors on a 200 keV FEI Tecnai Arctica microscope and a 300 keV FEI Tecnai G2 Polara microscope. A correction method was developed to correct for per-pixel differences in output. Timepix3 data were simulated for individual electron events using the package Geant4Medipix. Global statistical characteristics of the simulated detector response were in good agreement with experimental results. A convolutional neural network (CNN) was trained using the simulated data to predict the incident position of the electron within a pixel cluster. After training, the CNN predicted, on average, 0.50 pixel and 0.68 pixel from the incident electron position for 200 keV and 300 keV electrons respectively. The CNN improved the MTF of experimental data at half Nyquist from 0.39 to 0.70 at 200 keV, and from 0.06 to 0.65 at 300 keV respectively. We illustrate that the useful dose-lifetime of a protein can be measured within a 1 second exposure using Timepix3.

Thermal vacuum testing of Timepix3 detector

The thermal dependence of semiconductor detectors is one of their critical properties. This paper presents the results of the Timepix3 detector thermal vacuum testing, with respect to the effects on its properties and sensitivity. The Timepix3 represents a new generation of Timepix chips of the Medipix family, and it is equipped with an event-based mode of detection allowing for simultaneous measurement of the position, time and energy of an incident particle. Due to their properties, Timepix3 detectors are very suitable for space applications. Given that this is a relatively new device, the influence of temperature is not described in detail yet, especially for space usage. The operation of the device in a broad range of temperatures is required (e.q. QB50 mission on LEO from −20̂C to +50̂C). Timepix detectors have been used already in space missions, e.g. VZLUSAT-1, LUCID and SATRAM missions. In space, thermal cycling of the detectors occurs and this results in measurement distortion because both the noise edge and energy spectra are affected by changes in temperature. The experiments were performed on a detector equipped with a 300 μm thick Si sensor. The detector was equalised under various thermal conditions in vacuum and subsequently exposed to several energies of X-ray radiation corresponding to the characteristic radiation of 5 elements in the energy range of 4–24 keV. The results of these tests improve the knowledge regarding the behaviour of the essential components of the detector under extreme conditions. This new information can be used to improve measurements and thus minimise external influences, for example, in space applications but also in other fields where temperature stabilisation of the detector is very difficult or energy-consuming.

Study of Power Consumption of Timepix3 Detector

The Timepix3 readout chip—the latest member of the Medipix family of hybrid pixel detectors—brought several new functionalities in comparison with the older Timepix, i.e. a high hit-rate, a time granularity of 1.5625 ns, a data-driven readout scheme (with a per pixel dead time of approximately 475 ns), and the capability of measuring Time-over-Threshold (ToT) and Time-of-Arrival (ToA) in each pixel at the same time. However, the high power consumption of the Timepix3 in the standard setting prevents its use in applications with limited power budget. Moreover, the high power consumption poses the risk of overheating the sensor so that proper cooling is crucial. The presented work investigates the effect of different settings in the analogue and a digital part of the Timepix3 detector on its power consumption. Measurements were performed with the Timepix3 chipboard. The firmware of the Katherine readout was modified so that the user can monitor the power consumptions of analogue and digital part “on-line” (directly in the control software). In standard settings, a power consumption of approximately 1.3 W was found. By changes of internal DACs, the consumption could be reduced to 650 mW. Further reduction was achieved by the change of the clock management in the digital part of the Timepix3. In result, a power consumption of 216 mA could be achieved. In these low power settings, the ToA clock was reduced to 10 MHz and thus the time binning was 100 ns. The energy resolution was not affected significantly. The pixel dead time is also negatively affected when the matrix clock is reduced. In the case of 10 MHz, the minimal per pixel dead time is 1.9 μs.

Equalization of Medipix family detector energy thresholds using X-ray tube spectrum high energy cut-off

The Medipix detector family with single photon counting, photon energy discrimination and small pixel size makes it possible to carry out spectral X-ray imaging with high spatial resolution. In these detectors, the energy discrimination is performed independently in each pixel. For a good quality of spectral measurements, it is important that the globally defined energy threshold corresponds to the same energy in each pixel. The possibility of local adjustment of the pixel threshold is implemented by the chip design. In this paper, we present a method for thresholds equalization at an arbitrary energy using the X-ray tube spectrum high energy cut-off as a reference energy point. It is shown that the proposed method reduces the spread of pixel energy thresholds by more than 60 % in the energy range from 15 keV to 55 keV.

Katherine: Ethernet Embedded Readout Interface for Timepix3

The Timepix3—the latest generation of hybrid particle pixel detectors of Medipix family—yields a lot of new possibilities, i.e. a high hit-rate, a time resolution of 1.56 ns, event data-driven readout mode, and the capability of measuring the Time-over-Threshold (ToT - energy) and the Time-of-Arrival (ToA) simultaneously. This paper introduces a newly developed readout device for the Timepix3, called "Katherine", featuring a Gigabit Ethernet interface. The primary benefit of the Katherine is the operation of Timepix3 at long distance (up to 100 m) from computer or server, which is advantageous for the installation at beam lines, where the access is difficult or where radiation levels are too high for human interventions. The maximal hit-rate is limited by the bandwidth of the Ethernet connection (peer-to-peer connection; up to 16 Mhit/s). Since the Katherine interface is equipped with a processor of high computational power (ARM Cortex-A9 dual-core processor), it permits the use as a stand-alone (autonomous) radiation detector. The key features of the device are described in detail. These are the implemented high voltage power supply offering both polarities of bias voltage (up to ± 300 V), the automatic data sending to a sever via SSH, the automatic compensation of ToA values from columns with shifted matrix clock, etc. A dedicated control software was developed, which can be used for the detector preparation (sensor equalization, the DACs dependency scan, and the THL scan) and measurement control. Measured energy spectra from photon fields are shown.

A silicon detector in edge-on configuration for (spectral) Computed Tomography: proof of concept

This project focuses on a hybrid silicon pixel detector for Computed Tomography. In order to improve the attenuation efficiency of silicon for high energies, the active volume per unit area is increased by using the detector in edge-on configuration. In this geometry the sensor is illuminated from the side, with the pixel matrix parallel to the X-ray beam direction. Our setup consists of a 500 μm thick silicon sensor, bump-bonded to a chip from the Medipix family. Aim of the project is to test the feasibility of this geometry, finding its benefits and limitations. In particular, in this paper we show an important advantage of this configuration: energy discrimination along the detector depth. We propose a method to exploit this information, by including the beam hardening model both in the forward and in the backprojector of an iterative reconstruction algorithm. The first results, obtained on simulated data, show convergence and prove the feasibility of such an approach.

USB 3.0 readout and time-walk correction method for Timepix3 detector

The hybrid particle counting pixel detectors of Medipix family are well known. In this contribution we present new USB 3.0 based interface AdvaDAQ for Timepix3 detector. The AdvaDAQ interface is designed with a maximal emphasis to the flexibility. It is successor of FitPIX interface developed in IEAP CTU in Prague. Its modular architecture supports all Medipix/Timepix chips and all their different readout modes: Medipix2, Timepix (serial and parallel), Medipix3 and Timepix3. The high bandwidth of USB 3.0 permits readout of 1700 full frames per second with Timepix or 8 channel data acquisition from Timepix3 at frequency of 320 MHz. The control and data acquisition is integrated in a multiplatform PiXet software (MS Windows, Mac OS, Linux). In the second part of the publication a new method for correction of the time-walk effect in Timepix3 is described. Moreover, a fully spectroscopic X-ray imaging with Timepix3 detector operated in the ToT mode (Time-over-Threshold) is presented. It is shown that the AdvaDAQ's readout speed is sufficient to perform spectroscopic measurement at full intensity of radiographic setups equipped with nano- or micro-focus X-ray tubes.

Applications of the Medipix3-CT in combination with iterative reconstruction techniques

The pixelated semiconductor detectors of the Medipix family with their photon-counting abilities offer the possibility of high quality X-ray radiography as well as computed tomography. The generated signal from each photon is amplified and shaped before it is compared to an energy threshold. For a photon with an energy above the threshold the counter is incremented by one count. Photons below the operator-defined threshold do not increment the counter and therefore do not participate in the image formation. Furthermore, compared to other detectors like scintillators, an additional conversion step is dispensed due to the direct converting nature of photon-counting detectors, leading to a higher signal-to-noise-ratio. Additionally, the photon processing capabilities of photon-counting detectors allow photons to be weighted equally and not proportional to their energy as it is the case for charge integrating devices, where high energy photons are weighted stronger than low energy photons. Compared to integrating devices, this leads to an increase in contrast for images of both high and low contrast objects, hence improve object information. The use of photon-counting detectors in combination with iterative reconstruction techniques based on OSEM (ordered subset expectation maximization) algorithms is the basis of our computed tomography scans for material analysis. Due to its ability to operate with highly undersampled data sets, iterative reconstruction offers the possibility to decrease dose in CT scans. In order to identify the limits of the data set reduction, a first series of scans was performed to test, under real conditions, the CT-image quality when a strongly reduced amount of projections is used for reconstruction. In addition, the effect of a total variation minimization tool on these undersampled data sets was evaluated. Furthermore, this paper includes a number of recent CT-results with scans performed at two different setups within our facility.

Validation of Geant4 Pixel Detector Simulation Framework by Measurements With the Medipix Family Detectors

Monte Carlo simulations are an extensively used tool for developing and understanding radiation detector systems. In this work, we used results of several chips and readout modes of the Medipix detector family to validate a Geant4 based pixel detector framework, developed in our group, that is capable of simulating particle tracking, charge transport in the sensor material and different readout schemes. We experimentally verified the simulation with different detector geometries in terms of pixel pitch and size as well as sensor material and sensor thickness. The single pixel mode (SPM) and charge summing mode (CSM) in Medipix3 were evaluated with fluorescence and synchrotron radiation. The integration of the charge sensitive amplifier functionality in the simulation framework allowed to simulate the time-over-threshold mode of the Timepix chip. Simulation and measurement have been compared in terms of spectral resolution using threshold scans in photon counting mode (Medipix3) and time over threshold mode (Timepix). Further comparisons were done using X-ray tube spectra and beta decay to cover a broad energy range. Additionally, TCAD simulations are performed as a comparison to a well-established simulation method. The results show good agreement between simulation and measurement.

Edge-on illumination photon-counting for medical imaging

In medical X-ray Computed Tomography (CT) a silicon based sensor (300–1000 μm) in face-on configuration does not collect the incoming X-rays effectively because of their high energy (40–140 keV). For example, only 2% of the incoming photons at 100 keV are stopped by a 500 μm thick silicon layer. To increase the efficiency, one possibility is to use materials with higher Z (e.g. GaAs, CZT), which have some drawbacks compared to silicon, such as short carrier lifetime or low mobility. Therefore, we investigate whether illuminating silicon edge-on instead of face-on is a solution. Aim of the project is to find and take advantage of the benefits of this new geometry when used for a pixel detector. In particular, we employ a silicon hybrid pixel detector, which is read out by a chip from the Medipix family. Its capabilities to be energy selective will be a notable advantage in energy resolved (spectral) X-ray CT.

VeloPix: the pixel ASIC for the LHCb upgrade

The LHCb Vertex Detector (VELO) will be upgraded in 2018 along with the other subsystems of LHCb in order to enable full readout at 40 MHz, with the data fed directly to the software triggering algorithms. The upgraded VELO is a lightweight hybrid pixel detector operating in vacuum in close proximity to the LHC beams.The readout will be provided by a dedicated front-end ASIC, dubbed VeloPix, matched to the LHCb readout requirements and the 55 × 55 μm VELO pixel dimensions. The chip is closely related to the Timepix3, from the Medipix family of ASICs. The principal challenge that the chip has to meet is a hit rate of up to 900 Mhits/s, resulting in a required output bandwidth of more than 16 Gbit/s. The occupancy across the chip is also very non-uniform, and the radiation levels reach an integrated 400 Mrad over the lifetime of the detector.VeloPix is a binary pixel readout chip with a data driven readout, designed in 130 nm CMOS technology. The pixels are combined into groups of 2 × 4 super pixels, enabling a shared logic and a reduction of bandwidth due to combined address and time stamp information. The pixel hits are combined with other simultaneous hits in the same super pixel, time stamped, and immediately driven off-chip. The analog front-end must be sufficiently fast to accurately time stamp the data, with a small enough dead time to minimize data loss in the most occupied regions of the chip. The data is driven off chip with a custom designed high speed serialiser. The current status of the ASIC design, the chip architecture and the simulations will be described.

A Geant4 based framework for pixel detector simulation

The output from a hybrid pixel detector depends on the interaction of the radiation with the sensor material, the transport of the resulting charge in the sensor, the pulse processing in the readout circuit and processing of the resulting signal. In order to understand the full behaviour of the device and to predict the performance of future devices it is important to have a framework that can simulate the entire process in the detector system. Geant4 is a Monte Carlo based toolkit for simulation of particle interaction with matter which is developed and actively used for CERN experiments and detector development [1]. By extending the Monte Carlo code in Geant4 with a charge carrier transport model of the sensor material and basic amplifier functionality as well as read out logic, a simulation of the complete detector system is possible.The MEDIPIX is a state of the art hybrid pixel detector that allows bonding of a wide range of sensor materials [2,3]. Simulation models have been developed and tested for different chips from the MEDIPIX family. The simulation is defined using configuration files to set the geometry, sensor material properties, number of pixels, pixel pitch and chip properties. Source properties as well as filters and objects in the beam can be added for different experimental set-ups. The interaction of radiation with the sensor is taken into account in the transport of the charge carriers in the sensor material and a current induced in the pixel electrode that triggers an amplifier response. Simulation results have been verified with X-ray fluorescence and radioactive sources using MEDIPIX family chips. In this paper we present the developed simulation framework and first results.

Medipix3 CT for material sciences

Innovative detector systems for non-destructive material analysis and for medical diagnosis are an important development to improve the performance and the quality of examination methods. For a number of years now photon-counting X-ray detectors are being developed to process incoming X-ray photons as single events. These detectors facilitate a higher signal-to-noise ratio (SNR) than conventional, non-photon-counting, scintillator based detector systems, which detect X-ray photons indirectly through conversion into visible light. The Medipix is a pixelated photon counting semiconductor detector which features adjustable energy thresholds allowing energy selective, multispectral X-ray imaging. The Medipix chip is under continued development by the ``Medipix2 Collaboration'' and ``Medipix3 Collaboration'' at CERN [1]. The Medipix electronic offers 256 × 256 pixels with a pixel pitch of 55 × 55 μm2 and can be hybridized with different sensor materials like Si, CdTe or GaAs. The newest member of the Medipix family is the Medipix3 (ASIC in 0.13 μm CMOS technology) providing up to eight separate 12-bit counters per pixel. It offers a couple of different working modes [2], which are useful for X-ray imaging applications. A Medipix3 CT X-ray measuring station was built up for small animal X-ray imaging and non-destructive material analysis [3]. The combination of the low energy threshold ( ∼ 4 keV) of the Medipix3 with its multispectral capability enables tomographic investigations on objects with low absorption contrast. The advantage of photon counting, multispectral detectors like Medipix3 for material sciences will be presented here as well as a comparison with a scintillator based CT.

Measuring radiation environment in LHC or anywhere else, on your computer screen with Medipix

The Medipix family of chips use on-pixel pulse processing front-ends, digitization and counters to produce images of radiation. The devices have been derived from developments for the Large Hadron Collider (LHC) physics experiments at CERN. With the miniaturization of the associated readout system a new method of dosimetry becomes accessible, where single radiation quanta are detected and imaged. Several examples of dose measurements at highly differing dose rates are presented here: monitoring of background radiation on earth, in a flying airplane and in the ATLAS experiment at LHC. During proton collision runs as well as during the stops of the accelerator the dose can be measured, including characterization of different types of radiation. Thanks to the noiseless method of quantum imaging dosimetry, a large dynamic range can be achieved, employing only this single device. The dose rate extends from recording only a few quanta in hours, up to hundreds of quanta recorded in a fraction of a ms. With complementary methods for the analysis of the exposed image frames, one can cover 14 orders of magnitude.

Characterization of Medipix3 With Synchrotron Radiation

Medipix3 is the latest generation of photon counting readout chips of the Medipix family. With the same dimensions as Medipix2 (256 × 256 pixels of 55 μm × 55 μm pitch each), Medipix3 is however implemented in an 8-layer metallization 0.13 μm CMOS technology which leads to an increase in the functionality associated with each pixel over Medipix2. One of the new operational modes implemented in the front-end architecture is the Charge Summing Mode (CSM). This mode consists of a charge reconstruction and hit allocation algorithm which eliminates event-by-event the low energy counts produced by charge-shared events between adjacent pixels. The present work focuses on the study of the CSM mode and compares it to the Single Pixel Mode (SPM) which is the conventional readout method for these kind of detectors and it is also implemented in Medipix3. Tests of a Medipix3 chip bump-bonded to a 300 μm thick silicon photodiode sensor were performed at the Diamond Light Source synchrotron to evaluate the performance of the new Medipix chip. Studies showed that when Medipix3 is operated in CSM mode, it generates a single count per detected event and consequently the charge sharing effect between adjacent pixels is eliminated. However in CSM mode, it was also observed that an incorrect allocation of X-rays counts in the pixels occurred due to an unexpectedly high pixel-to-pixel threshold variation. The present experiment helped to better understand the CSM operating mode and to redesign the Medipix3 to overcome this pixel-to-pixel mismatch.

First CT using Medipix3 and the MARS-CT-3 spectral scanner

The MARS research group has created a new version of their scanner for taking improved spectral CT datasets. This version of the scanner (MARS-CT-3) has taken the first Medipix3 CT images of a phantom. MARS-CT-3 incorporates a new gantry, new x-ray sources and the new MARS readout board, as well as the ability to connect gas lines to the specimen. The new gantry has improved mechanical rigidity and can perform scans faster. Magnification can be controlled by moving the detector and the x-ray source independently. The brighter x-ray source means images can be taken six times faster. Gas lines allow the user to control various environmental factors inside the scanner, such as temperature, or deliver oxygen and anaesthetics, providing the ability to do a full spectroscopic CT scan of a live sedated biological specimen, such as a mouse. The new MARS readout is able to read from all current chips from the Medipix family, has faster image downloading, and the use of up to six Medipix detectors in parallel on the same chip carrier. The use of Medipix3 chips allows for compensation of charge sharing via Charge Summing Mode.

FITPix — fast interface for Timepix pixel detectors

The semiconductor pixel detector Timepix contains an array of 256 × 256 square pixels with pitch 55 μm. In addition to high spatial granularity the single quantum counting detector Timepix can provide also energy or time information in each pixel. This device is a powerful tool for radiation and particle detection, imaging and tracking. A new readout interface for silicon pixel detectors of the Medipix family has been developed in our group in order to provide a higher frame rate and enhanced flexibility of operation. The interface consists of a field programmable gate array, a USB 2.0 interface chip, DAC, ADC and a circuit which generates bias voltage for the sensor. The main control system is placed in the FPGA circuit which fully controls the Timepix device. This approach offers an easy way how to include new functionality and extended operation. The interface for Timepix supports all operation modes of the detector (counting, TOT, timing). The FITPix is a successor of the USB 1.22 Interface and the electronic readout is built with the latest available components, which allows achieving up to 90 frames per second with a single detector. The frame rate is about 20 times faster compared to the previous system while it maintains all same capabilities supported. In addition FITPix newly enables an adjustable clock frequency and hardware triggering which is a useful tool when there is the need for synchronized operation of multiple devices. Three modes of hardware trigger have been implemented: hardware trigger which starts the measurement, hardware trigger which terminates the measurement and hardware trigger which controls measurement fully. The entire system is fully powered through the USB bus. FITPix supports also readout from several detectors in chain in which case just an external power source is required. FITPix is a fully flexible device and the user needs no other equipment. FITPix combines high performance and mobility and it opens new fields of applications. The current version of the FITPix interface has dimension 45 mm × 60 mm.

Study of charge-sharing in MEDIPIX3 using a micro-focused synchrotron beam

X-ray photon-counting detectors consisting of a silicon pixel array sensor bump-bonded to a CMOS electronic readout chip offer several advantages over traditional X-ray detection technologies used for synchrotron applications. They offer high frame rate, dynamic range, count rate capability and signal-to-noise ratio. A survey of the requirements for future synchrotron detectors carried out at the Diamond Light Source synchrotron highlighted the needs for detectors with a pixel size of the order of 50μm. Reducing the pixel size leads to an increase of charge-sharing events between adjacent pixels and, therefore, to a degradation of the energy resolution and image quality of the detector. This effect was observed with MEDIPIX2, a photon-counting readout chip with a pixel size of 55μm. The lastest generation of the MEDIPIX family, MEDIPIX3, is designed to overcome this charge-sharing effect in an implemented readout operating mode referred to as Charge Summing Mode. MEDIPIX3 has the same pixel size as MEDIPIX2, but it is implemented in an 8-metal 0.13μm CMOS technology which enables increased functionality per pixel. The present work focuses on the study of the charge-sharing effect when the MEDIPIX3 is operated in Charge Summing Mode compared to the conventional readout mode, referred to as Single Pixel Mode. Tests of a standard silicon photodiode array bump-bonded to MEDIPIX3 were performed in beamline B16 at the Diamond Light Source synchrotron. A monochromatic micro-focused beam of 2.9μm x 2.2μm size at 15keV was used to scan a cluster of nine pixels in order to study the charge collection and X-ray count allocation process for each readout mode, Single Pixel Mode and Charge Summing Mode. The study showed that charge-shared events were eliminated when Medipix3 was operated in Charge Summing Mode.