Parallel extremal optimization in processor load balancing for distributed applications

The paper concerns parallel methods for extremal optimization (EO) applied in processor load balancing in execution of distributed programs. In these methods EO algorithms detect an optimized strategy of tasks migration leading to reduction of program execution time. We use an improved EO algorithm with guided state changes (EO-GS) that provides parallel search for next solution state during solution improvement based on some knowledge of the problem. The search is based on two-step stochastic selection using two fitness functions which account for computation and communication assessment of migration targets. Based on the improved EO-GS approach we propose and evaluate several versions of the parallelization methods of EO algorithms in the context of processor load balancing. Some of them use the crossover operation known in genetic algorithms. The quality of the proposed algorithms is evaluated by experiments with simulated load balancing in execution of distributed programs represented as macro data flow graphs. Load balancing based on so parallelized improved EO provides better convergence of the algorithm, smaller number of task migrations to be done and reduced execution time of applications.     

Single phase admission control for QoS-routing protocol in ad hoc networks

In this paper, in order to fulfill real-time traffic requirements in ad hoc networks, a novel and effective single phase admission control (SPAC) scheme for QoS-routing protocols has been proposed. The SPAC scheme is based on the ad hoc on-demand distance vector (AODV) protocol with slight modifications of control packets; network congestions are avoided by a simple and precise admission control that blocks most of the overloading flow requests in the route-discovery process. System simulations show that the SPAC scheme and its performance is comparable to the contention-aware admission control protocol (CACP)-Multihop performance in all respects; the SPAC scheme is also simpler. As compared to a QoS-aware routing protocol employing either “listen” or “hello” scheme, the SPAC protocol offers higher throughput and remarkably less end-to-end delays under heavy loads.     

Towards improving usage and management of supplies in healthcare: An ontology-based solution for sharing knowledge

The goal of this work is to contribute to an improvement in the management and usage of medical items in hospitals by developing an ontology-driven solution that organizes and describes clearly related knowledge. Experts in the purchasing and management of hospital supplies (administrative and clinical) were invited to participate at different stages of the ontology-based system development. As a result of the first stage, the HealthCatalog ontology was developed. This ontology models generic items included in a health catalog and their management. Secondly, a further refinement of the ontology was conducted by specifically studying the case of gloves. Twenty-seven references of different glove types were modeled and included in the ontology during the stage two. A prototype was then developed as a proof of concept and for the evaluation of the ontology. Finally, a usability evaluation was planned to improve the ontology and obtain feedback from experts after testing the system. Experts involved in the evaluation stressed its potential use in a real clinical environment and the benefits it would bring in terms of cost and sharing knowledge among clinical personnel. Our proposed ontology-based system provides an understandable and organized solution to capture knowledge regarding item management and usage. It addresses the integration challenge of health catalogs while providing a framework for collaborative sharing and knowledge acquisition among clinicians.     

Behavior monitoring under uncertainty using Bayesian surprise and optimal action selection

The increasing trend towards delegating tasks to autonomous artificial agents in safety–critical socio-technical systems makes monitoring an action selection policy of paramount importance. Agent behavior monitoring may profit from a stochastic specification of an optimal policy under uncertainty. A probabilistic monitoring approach is proposed to assess if an agent behavior (or policy) respects its specification. The desired policy is modeled by a prior distribution for state transitions in an optimally-controlled stochastic process. Bayesian surprise is defined as the Kullback–Leibler divergence between the state transition distribution for the observed behavior and the distribution for optimal action selection. To provide a sensitive on-line estimation of Bayesian surprise with small samples twin Gaussian processes are used. Timely detection of a deviant behavior or anomaly in an artificial pancreas highlights the sensitivity of Bayesian surprise to a meaningful discrepancy regarding the stochastic optimal policy when there exist excessive glycemic variability, sensor errors, controller ill-tuning and infusion pump malfunctioning. To reject outliers and leave out redundant information, on-line sparsification of data streams is proposed.     

创建有效的解决方案：从过程的角度出发

Global competition and declining margins have made enterprises in diverse industries increasingly aware that assuring low cost, product performance and high quality is no longer sufficient for long-term success. Integrating products and services to customized solutions can help firms to differentiate from their competitors. Based on 11 in-depth interviews with managers from solution providers and an exploratory survey with 45 solution providers we derive a four-step process of value creation as well as a set of critical activities and pitfalls within each step. Selling solutions require customer-supplier relational processes comprising analysis/consulting, design/configuration, implementation/delivery, and support/operation. To get a better understanding of the relational process we adopted this perspective on creating solutions in order to identify crucial routines and activities. Two main capabilities within this process can be identified: customer interaction and project management. Both capabilities are required in order to deliver more effective solutions at profitable prices.     

Automated detection of cyclic alternating pattern and classification of sleep stages using deep neural network

The visual sleep stages scoring by human experts is the current gold standard for sleep analysis. However, this method is tedious, time-consuming, prone to human errors, and unable to detect microstructure of sleep such as cyclic alternating pattern (CAP) which is an important diagnostic factor for the detection of sleep disorders such as insomnia and obstructive sleep apnea (OSA). The CAP is only observed as subtle changes in the electroencephalogram (EEG) signals during non-rapid eye movement (NREM) sleep, making it very difficult for human experts to discern. Hence, it is important to have an automated system developed using artificial intelligence for accurate and robust detection of CAP and sleep stages classification. In this study, a deep learning model based on 1-dimensional convolutional neural network (1D-CNN) is proposed for CAP detection and homogenous 3-class sleep stages classification, namely wakefulness (W), rapid eye movement (REM) and NREM sleep. The proposed model is developed using standardized EEG recordings. Our developed CNN network achieved good model performance for 3-class sleep stages classification with a classification accuracy of 90.46%. Our proposed model also yielded a classification accuracy of 73.64% using balanced CAP dataset, and sensitivity of 92.06% with unbalanced CAP dataset. Our proposed model correctly identified majority of A-phases which comprised of only 12.6% in the unbalanced dataset. The performance of the developed prototype is ready to be tested with more data before clinical application.

     

Fuzzy Sets for Feature Identification in Biomedical Signals with Self-Assessment of Reliability: An Adaptable Algorithm Modeling Human Procedure in BAEP Analysis

The analysis of biomedical signals has mostly been restricted to traditional signal-processing methods. This article proposes a different approach, applied to the evaluation of brainstem auditory evoked potentials. An automatic peak identification method is described which uses criteria applied by human evaluators in visual analysis. These criteria are defined as fuzzy sets and are combined using fuzzy operations, thus reflecting the weighting of different facts by humans. Membership values are interpreted as degrees of satisfaction which indicate the degree to which a sample satisfies a given criterion. The system judges its own performance in terms of degrees of reliability. Tests on a large set of clinical data showed high performance on good and average-quality curves. A substantial drawback was the assignment of too many peaks in poor potentials. The approaches presented here can easily be applied to similar one-dimensional (and higher) signal analysis tasks.