2D image reconstruction with a FPGA-based architecture in a gamma camera application

Abstract

A gamma camera is a device used to detect radioactivity in the form of gamma rays emitted by radioactive isotopes that have been introduced into the body as tracers. The position of the source of the radioactivity can be plotted and displayed to produce a two dimensional image for medical diagnostic. In this work a hardware architecture for reconstruction of 2D images based in FPGAs (Field Programmable Gate Arrays) in a gamma camera for small animals is presented. The hardware implementation takes the digitized voltage signals produced by the data acquisition electronics of the photomultiplier tubes and process them to allow identifying events and to determine the strength and positions of the interactions based on the logic of Anger to form a planar image. The architecture performs arithmetic operations, formats the data and stores them in memory blocks to be sent to the displaying stage. Through the use of parallelism techniques and an appropriate management of memory, the necessary logic to structure images is implemented inside the FPGA, reducing the time of processing to reach real time performance and to improve flexibility. The system has been implemented using a FPGA Virtex IV of the Xilinx's family obtaining an area occupation of around 22%, achieving an operation frequency of 322 MHz. These results show the potential of the hardware architecture implemented for its application in images reconstruction on line.