Développement d'une sonde intracérébrale à pixels actifs pour l'imagerie bêta du cerveau du rat libre de ses mouvements

Abstract : Over the last 20 years, many animal models have emerged, allowing the development of new approaches for the preclinical study of the healthy and pathological brain. Rodents have become key players in therapeutic advances. In this context, radioisotope imaging, which quantifies radioactive tracers with excellent sensitivity, is a prime tool for the study of brain processes in vivo. But so far, the most common radioimaging techniques require anesthesia or immobilization of the animal. However, anesthetics affect the biological processes studied. In addition, there is a keen interest in the simultaneous study of the behavior of the animal. The acquisition of a dynamic image of brain processes concomitant with the behavior of the awake and freely moving animal is valuable information for the study of addiction, memory, etc.At IMNC lab, we have approached behavioral neuroimaging with an original method based on intracerebral probes that measure the concentration of the radioactive tracer by direct detection of positrons in situ. The PIXSIC probe, based on a pixelized sensor with silicon diodes, demonstrated their relevance in the context of pharmacological studies with completely freely moving animals. However, PIXSIC has shown some limitations for its longitudinal use: a high level of noise due to electromagnetic perturbations, a high sensitivity to annihilation gamma radiation and a high mechanical fragility of the implant thinned to 200 microns.Based on the advent of CMOS technologies for the detection of charged particles in high energy physics, our ambition is to design MAPSSIC, a probe that responds to the difficulties highlighted by PIXSIC. CMOS sensors allows amplification at the pixel level, thus limiting electromagnetic noise. The sensitive volume can be reduced to a thickness of a few tens of microns, thus greatly reducing the sensitivity to gammas and allowing the increase of its total thickness to ensure its mechanical robustness. Finally, CMOS sensors allows us to design a highly pixelated detector to reach new imaging capabilities. This thesis aims to develop an optimized version of the probe. We imagined a first prototype CMOS sensor and we developed a Monte Carlo model to estimate its detection properties. We were able to show that his performances qualified it for the intended use, in terms of sensitivity, isoefficiency volume and deposited energy. We have also been able to explore several optimization parameters, the pixel dimensions and the thickness of the sensitive area, which allow us to consider MAPSSIC beyond the first prototype. With these theoretical bases we have produced several copies of the sensor. The developments that were established during the thesis then focused on a set of methodological tools, software and hardware to allow the physical characterization of the sensor using radioactive sources. We have been able to establish the uniformity of the pixel response and the event rate range ensuring the linearity of the count rate.These elements allowed us to conclude on the relevance of this sensor for the design of an autonomous imaging device. This consists of an implant made of two back-to-back sensors, an electronic system providing sensor control, signal reading and wireless communication and an acquisition station. In the context of the thesis, we have shown its suitability for the evaluation of the variations of the activity of a liquid beta+ radioactive source in which the implant has been immersed.
Complete list of metadatas

Contributor : Abes Star <>
Submitted on : Friday, February 8, 2019 - 12:36:08 PM
Last modification on : Tuesday, December 10, 2019 - 3:28:08 PM
Long-term archiving on: Thursday, May 9, 2019 - 2:55:14 PM


Version validated by the jury (STAR)


  • HAL Id : tel-02011986, version 1


Luis Ammour. Développement d'une sonde intracérébrale à pixels actifs pour l'imagerie bêta du cerveau du rat libre de ses mouvements. Médecine nucléaire. Université Paris-Saclay, 2018. Français. ⟨NNT : 2018SACLS497⟩. ⟨tel-02011986⟩



Record views


Files downloads