Corneliu T. C. Arsene

Partial Logistic Artificial Neural Network for Competing Risks Regularized With Automatic Relevance Determination

Time-to-event analysis is important in a wide range of applications from clinical prognosis to risk modeling for credit scoring and insurance. In risk modeling, it is sometimes required to make a simultaneous assessment of the hazard arising from two or more mutually exclusive factors. This paper applies to an existing neural network model for competing risks (PLANNCR), a Bayesian regularization with the standard approximation of the evidence to implement automatic relevance determination (PLANNCR-ARD). The theoretical framework for the model is described and its application is illustrated with reference to local and distal recurrence of breast cancer, using the data set of Veronesi (1995).

Decision support system for water distribution systems based on neural networks and graphs theory for leakage detection

This paper presents an efficient and effective decision support system (DSS) for operational monitoring and control of water distribution systems based on a three layer General Fuzzy Min-Max Neural Network (GFMMNN) and graph theory. The operational monitoring and control involves detection of pipe leakages. The training data for the GFMMNN is obtained through simulation of leakages in a water network for a 24h operational period. The training data generation scheme includes a simulator algorithm based on loop corrective flows equations, a Least Squares (LS) loop flows state estimator and a Confidence Limit Analysis (CLA) algorithm for uncertainty quantification entitled Error Maximization (EM) algorithm. These three numerical algorithms for modeling and simulation of water networks are based on loop corrective flows equations and graph theory. It is shown that the detection of leakages based on the training and testing of the GFMMNN with patterns of variation of nodal consumptions with or without confidence limits produces better recognition rates in comparison to the training based on patterns of nodal heads and pipe flows state estimates with or without confidence limits. It produces also comparable recognition rates to the original recognition system trained with patterns of data obtained with the LS nodal heads state estimator while being computationally superior by requiring a single architecture of the GFMMNN type and using a small number of pattern recognition hyperbox fuzzy sets built by the same GFMMNN architecture. In this case the GFMMNN relies on the ability of the LS loop flows state estimator of making full use of the pressure/nodal heads measurements existent in a water network.

A multi-platform comparison of efficient probabilistic methods in the prediction of total knee replacement mechanics

Explicit finite element (FE) and multi-body dynamics (MBD) models have been developed to evaluate total knee replacement (TKR) mechanics as a complement to experimental methods. In conjunction with these models, probabilistic methods have been implemented to predict performance bounds and identify important parameters, subject to uncertainty in component alignment and experimental conditions. Probabilistic methods, such as advanced mean value (AMV) and response surface method (RSM), provide an efficient alternative to the gold standard Monte Carlo simulation technique (MCST). The objective of the current study was to benchmark models from three platforms (two FE and one MBD) using various probabilistic methods by predicting the influence of alignment variability and experimental parameters on TKR mechanics in simulated gait. Predicted kinematics envelopes were on average about 2.6 mm for tibial anterior–posterior translation, 2.9° for tibial internal–external rotation and 1.9 MPa for tibial peak contact pressure for the various platforms and methods. Based on this good agreement with the MCST, the efficient probabilistic techniques may prove useful in the fast evaluation of new implant designs, including considerations of uncertainty, e.g. misalignment.

Control of autonomous robots using fuzzy logic controllers tuned by genetic algorithms

Truly autonomous vehicles will require both projective planning and reactive components in order to perform robustly. Projective components are needed for long term planning and re-planning where explicit reasoning about future states is required. Reactive components allow the system to always have some action available in real time, and themselves can exhibit robust behaviour, but lack the ability to explicitly reason about future states over a long time period. The paper emphasises creating the projective component but also offer a simple solution for reactive component. A genetic algorithm implements the projective component, which designs automatically a fuzzy logic controller by modifying the position of controller membership functions and the commands given to the robot. For the reactive component, a simple solution was adopted so that if the robot sensors detect new obstacles, the robot will try to move to a previous position.

Artificial Neural Networks Used in the Survival Analysis of Breast Cancer Patients: A Node-Negative Study

Artificial neural networks have been shown to be effective as general non-linear models with applications to medical diagnosis, prognosis and survival analysis. This chapter begins with a review of artificial neural networks used as non-linear regression models in the survival analysis of breast cancer patients. These techniques are of much interest because they allow modelling of time-dependent hazards in the presence of complex non-linear and non-additive effects between covariates. First, the role of neural networks is introduced within the context of statistical methods and parametric techniques for prognosis of survival in breast cancer. Second, these methods are applied in a study comprising node-negative breast cancer patients in order to evaluate the evidence for improved models or combination of prognostic indices to be used in a clinical environment. In particular, node-negative breast cancer is an early form of breast cancer in which cancer cells have not yet spread to the regional lymph nodes. There is much interest in determining the relevant prognostic factors that can allocate node-negative patients into prognostic groups correlating with the risk of disease relapse and mortality following surgery. This risk index can then be used to inform the choice of therapy. The Cox regression model and Artificial Neural Networks (ANN), a Partial Logistic Artificial Neural Network with Automatic Relevance Determination (PLANN-ARD) are used in order to identify and interpret the prognostic group allocation. A monthly retrospective cohort study with 5-year follow-up is conducted in pathologically node-negative patients selected from two datasets collected from Manchester Christie Hospital, UK.

Model selection with PLANN-CR-ARD

This paper presents a new compensation mechanism to be used with a Partial Logistic Artificial Neural Network for Competing Risks with Automatic Relevance Determination (PLANN-CR-ARD) and tested comprehensibly on a real breast cancer dataset with excellent convergence properties and numerical stability for the non-linear model. The Model Selection is implemented for the PLANNCRARD model, benefiting from a scaling of the prior error term which together with the data error term forms the total error function that is optimized. The PLANN-CR-ARD proves to be an excellent prognostic tool that can be used in regression analysis tasks such as the survival analysis of cancer datasets.

PLANN-CR-ARD model predictions and Non-parametric estimates with Confidence Intervals

This paper investigates the performance of the PLANN-CR-ARD network predictions through a comparison with the confidence intervals and the non-parametric estimates obtained from the survival analysis of a Primary Billiary Cirrhosis (PBC) dataset. The predictions of the PLANN-CR-ARD model are marginalized using two methods: approximation of the integral of marginalization and the Monte Carlo method. The numerical results show that the PLANN-CR-ARD predicts very well, the results being situated within the confidence intervals of the non-parametric estimates. The Model Selection is also performed on the same dataset. The PLANN-CR-ARD can be used to explore the non-linear interdependencies between the predicted outputs and the input data which in survival analysis describes the characteristics of the patients.

Confidence Limit Analysis of Water Distribution Systems Based on a Least Squares Loop Flows State Estimation Technique

This paper presents a novel algorithm for uncertainty quantification in water distribution systems, which is termed also Confidence Limit Analysis (CLA), in the context of a Least Squares (LS) state estimator based on the loop corrective flows and the variation of nodal demands as state variables. The confidence limits predicted with the novel algorithm called Error Maximization (EM) method are evaluated with respect to two other more established CLA algorithms based on an Experimental Sensitivity Matrix (ESM) and on a sensitivity matrix obtained with the LS nodal heads equations state estimator. The predicted confidence limits show that the novel EM algorithm is comparable to the other CLA algorithms shown in the paper and due to its computational efficiency renders it suitable for online decision support systems for water distribution systems.

A Study on Modeling and Simulation of Water Distribution Systems Based on Loop Corrective Flows and Containing Controlling Hydraulics Elements

In the context of water networks, in order to supply water to consumers without any disruption in service, the state of the system has to be monitored. One way to achieve this is by using simulators algorithms that provide means of combining the water consumptions called also pseudo-measurements by relating them to the mathematical model of the system. Therefore a comprehensive description of the water distribution system should be used to attain a true picture of the state of the system. This implies that all hydraulics elements including non-linear controlling elements such as pumps and valves should be included in the description of the water distribution system. This paper addresses the problem of simulation of water networks containing controlling hydraulics elements by using simulator algorithms constructed with the loop corrective flows equations. It is well known that these controlling hydraulics elements may cause numerical difficulties in the simulations due to their discontinuous properties. It is shown that using the loop equations can provide an adequate mean for handling the discontinuous nature of the controlling hydraulics elements. This is in spite of having to deal with the loop structure of the water network in the simulator algorithm which implies the use of graph theory. The case of non-linear elements check valves, pumps and pressure reducing valves is presented. The main contribution of the paper is that it proposes a partial rebuilding of the spanning tree obtained with the Depth First (DF) search, which partial rebuilding results from the changing of the status of the non-linear controlling hydraulics elements during the convergence of the Newton-Raphson numerical optimization method used by the simulator algorithm. The numerical results are verified with the EPANET software which is an established computer program used in the simulation of water networks.

Bayesian Neural Network Applied in Medical Survival Analysis of Primary Biliary Cirrhosis

A benchmark medical study is realized for a Primary Biliary Cirrhosis (PBC) dataset by using two different versions of a Bayesian Neural Network (BNN) entitled Partial Logistic Artificial Neural Network for Competing Risks with Automatic Relevance Determination (PLANN-CR-ARD). The two BNN versions are based on two different compensation mechanisms which are designed to preserve the numerical stability of the PLANN-CR-ARD model and to calculate the marginalized network results. The predictions of the PLANN-CR-ARD models are comparable to the non-parametric estimates obtained through the survival analysis of the PBC dataset. The input variables from the PBC dataset which can have a strong influence on the outcome of the disease are determined. The PLANN-CR-ARD models can be used to investigate the non-linear inter-dependencies between the predicted outputs and the input data which consist of the characteristics of the PBC patients.