An information-theoretic approach to estimating ultrasound backscatter characteristics.

Analysis of backscatter in the ultrasound echo envelope, in conjunction with ultrasound B-scans, can provide important information for tissue characterization and pathology diagnosis. Statistical models have often proven useful in modeling backscatter. In this paper, an innovative approach to backscatter analysis based on generalized entropies and neural function approximation is presented. Entropy measures are shown to provide accurate estimates of scatterer density, regularity, and SNR of the amplitude distribution. Specific scattering distributions need not be assumed. Experimental results on ground truth envelopes show that generalized entropies can be used to accurately estimate backscatter properties.

A new design method for the complex-valued multistate Hopfield associative memory.

A method to store each element of an integral memory set M subset {1,2,...,K}/sup n/ as a fixed point into a complex-valued multistate Hopfield network is introduced. The method employs a set of inequalities to render each memory pattern as a strict local minimum of a quadratic energy landscape. Based on the solution of this system, it gives a recurrent network of n multistate neurons with complex and symmetric synaptic weights, which operates on the finite state space {1,2,...,K}/sup n/ to minimize this quadratic functional. Maximum number of integral vectors that can be embedded into the energy landscape of the network by this method is investigated by computer experiments. This paper also enlightens the performance of the proposed method in reconstructing noisy gray-scale images.

Extraction of rules from artificial neural networks for nonlinear regression.

Neural networks (NNs) have been successfully applied to solve a variety of application problems including classification and function approximation. They are especially useful as function approximators because they do not require prior knowledge of the input data distribution and they have been shown to be universal approximators. In many applications, it is desirable to extract knowledge that can explain how Me problems are solved by the networks. Most existing approaches have focused on extracting symbolic rules for classification. Few methods have been devised to extract rules from trained NNs for regression. This article presents an approach for extracting rules from trained NNs for regression. Each rule in the extracted rule set corresponds to a subregion of the input space and a linear function involving the relevant input attributes of the data approximates the network output for all data samples in this subregion. Extensive experimental results on 32 benchmark data sets demonstrate the effectiveness of the proposed approach in generating accurate regression rules.

Bi-directional computing architecture for time series prediction.

A number of neural network models and training procedures for time series prediction have been proposed in the technical literature. These models studied for different time-variant data sets have typically used uni-directional computation flow or its modifications. In this study, on the contrary, the concept of bi-directional computational style is proposed and applied to prediction tasks. A bi-directional neural network model consists of two subnetworks performing two types of signal transformations bi-directionally. The networks also receive complementary signals from each other through mutual connections. The model not only deals with the conventional future prediction task, but also with the past prediction, an additional task from the viewpoint of the conventional approach. An improvement of the performance is achieved through making use of the future-past information integration. Since the coupling effects help the proposed model improve its performance, it is found that the prediction score is better than with the traditional uni-directional method. The bi-directional predicting architecture has been found to perform better than the conventional one when tested with standard benchmark sunspots data.

Nonlinear blind source separation using a radial basis function network.

This paper proposes a novel neural-network approach to blind source separation in nonlinear mixture. The approach utilizes a radial basis function (RBF) neural-network to approximate the inverse of the nonlinear mixing mapping which is assumed to exist and able to be approximated using an RBF network. A contrast function which consists of the mutual information and partial moments of the outputs of the separation system, is defined to separate the nonlinear mixture. The minimization of the contrast function results in the independence of the outputs with desirable moments such that the original sources are separated properly. Two learning algorithms for the parametric RBF network are developed by using the stochastic gradient descent method and an unsupervised clustering method. By virtue of the RBF neural network, this proposed approach takes advantage of high learning convergence rate of weights in the hidden layer and output layer, natural unsupervised learning characteristics, modular structure, and universal approximation capability. Simulation results are presented to demonstrate the feasibility, robustness, and computability of the proposed method.

Blind extraction of singularly mixed source signals.

This paper introduces a novel technique for sequential blind extraction of singularly mixed sources. First, a neural-network model and an adaptive algorithm for single-source blind extraction are introduced. Next, extractability analysis is presented for singular mixing matrix, and two sets of necessary and sufficient extractability conditions are derived. The adaptive algorithm and neural-network model for sequential blind extraction are then presented. The stability of the algorithm is discussed. Simulation results are presented to illustrate the validity of the adaptive algorithm and the stability analysis. The proposed algorithm is suitable for the case of nonsingular mixing matrix as well as for singular mixing matrix.

A dynamical system perspective of structural learning with forgetting.

Structural learning with forgetting is an established method of using Laplace regularization to generate skeletal artificial neural networks. In this paper we develop a continuous dynamical system model of regularization in which the associated regularization parameter is generalized to be a time-varying function. Analytic results are obtained for a Laplace regularizer and a quadratic error surface by solving a different linear system in each region of the weight space. This model also enables a comparison of Laplace and Gaussian regularization. Both of these regularizers have a greater effect in weight space directions which are less important for minimization of a quadratic error function. However, for the Gaussian regularizer, the regularization parameter modifies the associated linear system eigenvalues, in contrast to its function as a control input in the Laplace case. This difference provides additional evidence for the superiority of the Laplace over the Gaussian regularizer.

Classification algorithms for quantitative tissue characterization of diffuse liver disease from ultrasound images.

Visual criteria for diagnosing diffused liver diseases from ultrasound images can be assisted by computerized tissue classification. Feature extraction algorithms are proposed in this paper to extract the tissue characterization parameters from liver images. The resulting parameter set is further processed to obtain the minimum number of parameters which represent the most discriminating pattern space for classification. This preprocessing step has been applied to over 120 distinct pathology-investigated cases to obtain the learning data for classification. The extracted features are divided into independent training and test sets, and are used to develop and compare both statistical and neural classifiers. The optimal criteria for these classifiers are set to have minimum classification error, ease of implementation and learning, and the flexibility for future modifications. Various algorithms of classification based on statistical and neural network methods are presented and tested. The authors show that very good diagnostic rates can be obtained using unconventional classifiers trained on actual patient data.

Complex-valued multistate neural associative memory.

A model of a multivalued associative memory is presented. This memory has the form of a fully connected attractor neural network composed of multistate complex-valued neurons. Such a network is able to perform the task of storing and recalling gray-scale images. It is also shown that the complex-valued fully connected neural network may be considered as a generalization of a Hopfield network containing real-valued neurons. A computational energy function is introduced and evaluated in order to prove network stability for asynchronous dynamics. Storage capacity as related to the number of accessible neuron states is also estimated.

Neural network predicted peak and trough gentamicin concentrations.

Predictions of steady state peak and trough serum gentamicin concentrations were compared between a traditional population kinetic method using the computer program NONMEM to an empirical approach using neural networks. Predictions were made in 111 patients with peak concentrations between 2.5 and 6.0 micrograms/ml using the patient factors age, height, weight, dose, dose interval, body surface area, serum creatinine, and creatinine clearance. Predictions were also made on 33 observations that were outside the 2.5 and 6.0 micrograms/ml range. Neural networks made peak serum concentration predictions within the 2.5-6.0 micrograms/ml range with statistically less bias and comparable precision with paired NONMEM predictions. Trough serum concentration predictions were similar using both neural networks and NONMEM. The prediction error for peak serum concentrations averaged 16.5% for the neural networks and 18.6% for NONMEM. Average prediction errors for serum trough concentrations were 48.3% for neural networks and 59.0% for NONMEM. NONMEM provided numerically more precise and less biased predictions when extrapolating outside the 2.5 and 6.0 micrograms/ml range. The observed peak serum concentration distribution was multimodal and the neural network reproduced this distribution with less difference between the actual distribution and the predicted distribution than NONMEM. It is concluded that neural networks can predict serum drug concentrations of gentamicin. Neural networks may be useful in predicting the clinical pharmacokinetics of drugs.

Sufficient condition for convergence of a relaxation algorithm in actual single-layer neural networks.

Application of the contraction mapping theorem to single-layer feedback neural networks of a gradient-type is discussed. The sufficient condition for stability of a relaxation algorithm in actual continuous-time networks is derived and illustrated with an example. Results showing the stability of a numerical solution obtained with the relaxation algorithm are presented.