In the context of the development of an in-beam time-of-flight positron emission tomography demonstrator dedicated to the in vivo monitoring of delivered dose in hadron-therapy, we evaluate the potential performance of front-end architectures based on sampling and digital pulse processing to reconstruct the energy and time of event. In this paper, we evaluate the requirements for sampling frequency and analog-to-digital converter (ADC) resolution. The timing algorithm is a digital adaptation of the constant fraction discriminator principle, with a step of interpolation using a low-pass filter based on the cubic spline technique. We demonstrate the interest of the interpolation to lower the sampling frequency requirement, by improving the signal reconstruction compared with a simple linear interpolation. Experimental tests were performed on pulse libraries acquired from a set up composed of LYSO and LaBr3 scintillators coupled to H6533 photomultipliers, and a sampling oscilloscope operating at 10 GHz. By offline processing of the signals with variable parameters (initial frequency, interpolator bandwidth, resampling frequency, and ADC resolution), we examined the impact of these parameters on time resolution. Results for the tested detectors suggest a minimal required sampling rate of 1.5 GHz, while the ADC resolution can be as small as 5 b. A logarithmic ADC could be more efficient, with a strict minimum of 4 b.

2012 is the penultimate year of financial support by the CPER 2007-2013 for ETOILE's research program, sustained by the PRRH at the University Claude Bernard. As with each edition we make the annual review of the research in this group, so active for over 12 years now. Over the difficulties in the decision-making process for the implementation of the ETOILE Center, towards which all our efforts are focussed, some themes (work packages) were strengthened, others have progressed, or have been dropped. This is the case of the eighth theme (technological developments), centered around the technology for rotative beam distribution heads (gantries) and, after being synchronized with the developments of ULICE's WP6, remained so by ceasing its activities, coinciding also with the retirement of its historic leader at IPNL, Marcel Bajard. Topic number 5 (In silico simulations) has suffered the departure of its leader, Benjamin Ribba, although the work has still been provided by Branka Bernard, a former postdoctoral fellow in Lyon Sud, and now back home in Croatia, still in contract with UCBL for the ULICE project. Aside from these two issues (and the fact that the theme Medico-economical simulations is now directly linked to the first one (Medical Project), the rest of the teams are growing, as evidenced by the publication statistics at the beginning of this report. This is obviously due to the financial support of our always faithful regional institutions, but also to the synergy that the previous years, the European projects, the arrival of the PRIMES LabEx, and the national France Hadron infrastructure have managed to impulse. The Rhone-Alpes hadron team, which naturally includes the researchers of LPC at Clermont, should also see its influence result in a strong presence in France Hadron's regional node, which is being organized. The future of this regional research is not yet fully guaranteed, especially in the still uncertain context of ETOILE, but the tracks are beginning to emerge to allow past and present efforts translate into a long future that we all want to see established. Each of the researchers in PRRH is aware that 2013 will be (and already is) the year of great challenge : for ETOILE, for the PRRH, for hadron therapy in France, for French hadrontherapy in Europe (after the opening and beginning of treatments in the German [HIT Heidelberg, Marburg], Italian [CNAO, Pavia] and Austrian [MedAustron, Wien Neuerstadt]) centers. Let us meet again in early 2014 for a comprehensive review of the past and a perspective for the future ...

With the development of fast sampling electronics, digital pulse processing techniques for PET signals are raising interest. The optimal filter (OF) algorithm reconstructs pulse amplitude and time by two weighted sums, making it compatible with real-time implementation. The filters are usually optimized for stationary noise. We developed and tested a method to optimize the filters for the nonstationary noise of scintillation pulses. It is based on offline statistical analysis of coincident waveforms that could be applied during the system initialization phase. Experimental tests were done on a coincidence setup with two detection blocks composed of a fast inorganic scintillator ( LaBr3 or LYSO) coupled to a photodetector (APD or PMT), preamplifiers and prefilters. The signals were sampled at high rate (250 MHz for APDs, 5 GHz for PMTs) and treated offline. The optimization of the filter coefficients for nonstationary noise yielded a significant improvement compared to those optimized for stationary noise, resp. 368 ps and 632 ps fwhm in coincidence for the LYSO-PMT setup. However, little improvement was achieved compared to leading-edge (DLED) and constant fraction (DCFD) discriminator algorithms (resp. 419 ps, 435 ps fwhm). Indeed, the adjustment of thresholds can be interpreted as an optimization for nonstationary noise. Yet, OF is more robust to white noise than DLED or DCFD. The applicability to PET is discussed.

We tested a digital front-end concept in order to evaluate the time resolution of PET detectors based on APD or PMT with digital read-out. Measurements were done on a coincidence set-up with two detection blocks composed of a fast inorganic scintillator (LaBr3 or LYSO) coupled to a photodetector (APD or PMT), preamplificators and prefilters. The signals were sampled at high rate (250MHz for APDs, 5GHz for PMTs) and treated offline. Two different timing algorithms were applied: a digital method deriving from constant fraction discriminator, and an optimal filtering technique based on parameter estimation with minimal variance. The classical optimal filter was adapted to the non-stationary noise conditions, with a significant improvement of timing resolution. We describe these algorithms and discuss their performances.

This paper describes the Analog to Digital Converter developed for the front end electronic of the IN2P3 INNOTEP project by the “pole microelectronique Rhone-Auvergne”. (Collaboration between LPC Clermont-Ferrand and IPNL Lyon). This ADC is a 4 stages 2.5 bits per stage pipe line with open loops track and holds and amplifiers. It runs at 100MSamples/s and has 8 bits resolution. The stages used two lines, the gain line and the comparison line, with most operators running in current. The main idea of this current line is to make a first step toward an all in current structure. Currently, this ADC is designed with a 0,35µm SiGe technology.

The INNOTEP collaboration investigates for a technological adaptation of High Energy Physics (HEP) detection and acquisition concepts to the PET issues. We present preliminary results of innovative electronics architecture for PET.The concept is based on a deadtimeless pipelined processing of the photosensors signals. After shaping and sampling by a free-running ADC, the pulses are digitally filtered to extract time and energy. The data is processed and selected inline before storage.We present the first tests of the low noise front-end electronics - preamplifier, shaper, ADC - and a Time Of Flight-capable digital filtering technique.

The purpose of this study was to show how advanced concepts of compact, lossless and ”Time Of Flight” capable electronics similar to those foreseen for the LHC and ILC experiments could be fairly and easily transferred to the medical imaging field through Positron Emission Tomography scanners. As a wish of explanation, the two overriding weaknesses of PET camera readout electronics, namely timing resolution and dead-time, were investigated analytically and with the help of a Monte-Carlo simulator presently dedicated to this task. Figures have shown there was rather space available for count rate enhancement, especially through a huge decrease of the timing resolution well below the nanosecond. A solution retained and proposed here for the electronics has been partly drawn from the long experience led in High Energy Physics where this last requirement is compulsory. Also appreciable, thanks to its structure entirely pipelined, this scheme enables problems of dead time to be overcome.