Maximum entropy co-processor for computed tomography

Author

Chang, Steven ; Peckerar, M. ; Marrian, Christie

Author_Institution

Dept. of Defense, Fort Meade, MD, USA

fYear

1994

fDate

1-4 May 1994

Firstpage

343

Lastpage

346

Abstract

In this paper, we present a neural net co-processor capable of performing computed tomographic image reconstruction. The circuit performs the Radon transformation using a cost function gradient descent method. The unique aspect of this co-processor is the incorporation of informational entropy as a regularizer in the optimization problem. A 10 pixel×10 pixel array was designed and fabricated in 2 μm CMOS technology. Convergence time of the array was less than 5 μs. Issues relating to scaling the array to larger sizes are discussed in this paper

Keywords

CMOS analogue integrated circuits; Radon transforms; VLSI; analogue processing circuits; computerised tomography; convergence; coprocessors; image processing equipment; image reconstruction; maximum entropy methods; neural chips; optimisation; real-time systems; 10 pixel; 100 pixel; 2 micron; 5 mus; CMOS technology; Radon transformation; computed tomographic image reconstruction; computerised tomography; convergence time; cost function gradient descent method; maximum entropy coprocessor; neural net co-processor; optimization problem; scaling; CMOS technology; Circuits; Computed tomography; Coprocessors; Entropy; Equations; Image reconstruction; Neural networks; Pixel; X-ray imaging;

fLanguage

English

Publisher

ieee

Conference_Titel

Custom Integrated Circuits Conference, 1994., Proceedings of the IEEE 1994

Conference_Location

San Diego, CA

Print_ISBN

0-7803-1886-2

Type

conf

DOI

10.1109/CICC.1994.379705

Filename

379705