What size net gives valid generalization?
Neural Computation
Learning internal representations by error propagation
Parallel distributed processing: explorations in the microstructure of cognition, vol. 1
Artificial convolution neural network for medical image pattern recognition
Neural Networks - Special issue: automatic target recognition
Hi-index | 0.00 |
We have developed a computerized method using a neuralnetwork for the segmentation of lung fields in chest radiography. The lung is the primary region of interest in routine chestradiography diagnosis. Since computer is expected to performdisease pattern search automatically, it is important to designappropriate algorithms to delineate the region of interest. Areliable segmentation method is essential to facilitate subsequentsearches for image patterns associated with lung diseases. In thisstudy, we employed a shift invariant neural network coupled witherror back-propagation training method to extract the lungfields. A set of computer algorithms were alsodeveloped for smoothing the initially detected edges of lungfields. Our preliminary results indicated that 86% of thesegmented lung fields globally matched the original chestradiographs. We also found that the method facilitates the development of computer algorithms in the field of computer-aideddiagnosis.