Title :
SPECT reconstruction using a backpropagation neural network implemented on a massively parallel SIMD computer
Author :
Kerr, John P. ; Barlett, E.B.
Author_Institution :
Iowa State Univ., Ames, IA, USA
Abstract :
The feasibility of reconstructing a single photon emission computed tomography (SPECT) image via the parallel implementation of a backpropagation neural network is shown. The MasPar MP-1 is a single-instruction multiple-data (SIMD) massively parallel machine, composed of a 128×128 array of 4-bit processors. The neural network is distributed on the array by dedicating a processor to each node and each interconnection of the network. An 8×8 SPECT image slice section is projected into eight planes. It is shown that, based on the projections, the neural network can produce the original SPECT slice image exactly. Likewise, when trained on two parallel slices, separated by one slice, the neural network is able to reproduce the center, untrained image to an RMS (root mean square) error of 0.001928
Keywords :
backpropagation; computerised tomography; diagnostic radiography; image reconstruction; medical image processing; neural nets; parallel processing; 4 bit; MasPar MP-1; SPECT reconstruction; SPECT slice image; backpropagation neural network; massively parallel SIMD computer; single photon emission computed tomography; Application software; Artificial neural networks; Backpropagation; Computer networks; Concurrent computing; Electrical capacitance tomography; Image reconstruction; Neural networks; Optical computing; Single photon emission computed tomography;
Conference_Titel :
Computer-Based Medical Systems, 1992. Proceedings., Fifth Annual IEEE Symposium on
Conference_Location :
Durham, NC
Print_ISBN :
0-8186-2742-5
DOI :
10.1109/CBMS.1992.245026