High-speed reconstruction of spect images with a tailored piecewise neural network
Conference
·
OSTI ID:206451
Artificial neural networks (ANNs) have proven to be highly adept at mapping complex functional relationships. We have previously shown that a standard backpropagation neural network can be trained to reconstruct sections of single photon emission computed tomography (SPECT) images based on the planar image projections as inputs. In this study, we demonstrate that a neural network that utilizes a tailored three-phase piecewise activation function is able to perform high-speed reconstructions of SPECT images after learning the relationship between the planar images and the tomographic reconstructions. In addition, the tailored piecewise neural network produces reconstructions with significantly lower RMS error, and does so in far less training iterations, than a standard backpropagation ANN. The tailored piecewise function used in this research enables the network to train on a continuous range of outputs more efficiently than with a standard sigmoidal function. Based on the results obtained, we hypothesize that the optimal ANN transfer function or functions, are directly related to the statistical distribution of the training set data. As a preliminary demonstration, a neural network with statistically derived activation functions is shown to have better training and generalization characteristics for SPECT reconstruction than either the single sigmoidal or the three- phase sigmoidal activation functions.
- Research Organization:
- Iowa State Univ. of Science and Technology, Ames, IA (United States)
- Sponsoring Organization:
- USDOE, Washington, DC (United States)
- DOE Contract Number:
- FG02-92ER75700
- OSTI ID:
- 206451
- Report Number(s):
- CONF-931051--3; ON: DE96007022
- Country of Publication:
- United States
- Language:
- English
Similar Records
The parallel implementation of a backpropagation neural network and its applicability to SPECT image reconstruction
The parallel implementation of a backpropagation neural network and its applicability to SPECT image reconstruction
SPECT reconstruction using a backpropagation neural network implemented on a massively parallel SIMD computer
Thesis/Dissertation
·
Tue Dec 31 23:00:00 EST 1991
·
OSTI ID:10138858
The parallel implementation of a backpropagation neural network and its applicability to SPECT image reconstruction
Technical Report
·
Tue Dec 31 23:00:00 EST 1991
·
OSTI ID:6879460
SPECT reconstruction using a backpropagation neural network implemented on a massively parallel SIMD computer
Conference
·
Wed Dec 30 23:00:00 EST 1992
·
OSTI ID:212515