Automated nonlinear system modelling with multiple neural networks

This article discusses the identification of nonlinear dynamic systems using multi-layer perceptrons (MLPs). It focuses on both structure uncertainty and parameter uncertainty, which have been widely explored in the literature of nonlinear system identification. The main contribution is that an integrated analytic framework is proposed for automated neural network structure selection, parameter identification and hysteresis network switching with guaranteed neural identification performance. First, an automated network structure selection procedure is proposed within a fixed time interval for a given network construction criterion. Then, the network parameter updating algorithm is proposed with guaranteed bounded identification error. To cope with structure uncertainty, a hysteresis strategy is proposed to enable neural identifier switching with guaranteed network performance along the switching process. Both theoretic analysis and a simulation example show the efficacy of the proposed method.

Automated versus non-automated weaning for reducing the duration of mechanical ventilation for critically ill adults and children (Review)

Abstract
Background: Automated closed loop systems may improve adaptation of the mechanical support to a patient's ventilatory needs and
facilitate systematic and early recognition of their ability to breathe spontaneously and the potential for discontinuation of
ventilation.

Objectives: To compare the duration of weaning from mechanical ventilation for critically ill ventilated adults and children when managed
with automated closed loop systems versus non-automated strategies. Secondary objectives were to determine differences
in duration of ventilation, intensive care unit (ICU) and hospital length of stay (LOS), mortality, and adverse events.

Search methods: We searched the Cochrane Central Register of Controlled Trials (CENTRAL) (The Cochrane Library 2011, Issue 2); MEDLINE (OvidSP) (1948 to August 2011); EMBASE (OvidSP) (1980 to August 2011); CINAHL (EBSCOhost) (1982 to August 2011); and the Latin American and Caribbean Health Sciences Literature (LILACS). In addition we received and reviewed auto-alerts for our search strategy in MEDLINE, EMBASE, and CINAHL up to August 2012. Relevant published reviews were sought using the Database of Abstracts of Reviews of Effects (DARE) and the Health Technology Assessment Database (HTA Database). We also searched the Web of Science Proceedings; conference proceedings; trial registration websites; and reference lists of relevant articles.

Selection criteria: We included randomized controlled trials comparing automated closed loop ventilator applications to non-automated weaning
strategies including non-protocolized usual care and protocolized weaning in patients over four weeks of age receiving invasive mechanical ventilation in an intensive care unit (ICU).

Data collection and analysis: Two authors independently extracted study data and assessed risk of bias. We combined data into forest plots using random-effects modelling. Subgroup and sensitivity analyses were conducted according to a priori criteria.

Main results: Pooled data from 15 eligible trials (14 adult, one paediatric) totalling 1173 participants (1143 adults, 30 children) indicated that automated closed loop systems reduced the geometric mean duration of weaning by 32% (95% CI 19% to 46%, P =0.002), however heterogeneity was substantial (I2 = 89%, P < 0.00001). Reduced weaning duration was found with mixed or
medical ICU populations (43%, 95% CI 8% to 65%, P = 0.02) and Smartcare/PS™ (31%, 95% CI 7% to 49%, P = 0.02) but not in surgical populations or using other systems. Automated closed loop systems reduced the duration of ventilation (17%, 95% CI 8% to 26%) and ICU length of stay (LOS) (11%, 95% CI 0% to 21%). There was no difference in mortality rates or hospital LOS. Overall the quality of evidence was high with the majority of trials rated as low risk.

Authors' conclusions: Automated closed loop systems may result in reduced duration of weaning, ventilation, and ICU stay. Reductions are more
likely to occur in mixed or medical ICU populations. Due to the lack of, or limited, evidence on automated systems other than Smartcare/PS™ and Adaptive Support Ventilation no conclusions can be drawn regarding their influence on these outcomes. Due to substantial heterogeneity in trials there is a need for an adequately powered, high quality, multi-centre randomized
controlled trial in adults that excludes 'simple to wean' patients. There is a pressing need for further technological development and research in the paediatric population.

Robust automated tumour segmentation on histological and immunohistochemical tissue images

Tissue microarray (TMA) is a high throughput analysis tool to identify new diagnostic and prognostic markers in human cancers. However, standard automated method in tumour detection on both routine histochemical and immunohistochemistry (IHC) images is under developed. This paper presents a robust automated tumour cell segmentation model which can be applied to both routine histochemical tissue slides and IHC slides and deal with finer pixel-based segmentation in comparison with blob or area based segmentation by existing approaches. The presented technique greatly improves the process of TMA construction and plays an important role in automated IHC quantification in biomarker analysis where excluding stroma areas is critical. With the finest pixel-based evaluation (instead of area-based or object-based), the experimental results show that the proposed method is able to achieve 80% accuracy and 78% accuracy in two different types of pathological virtual slides, i.e., routine histochemical H&E and IHC images, respectively. The presented technique greatly reduces labor-intensive workloads for pathologists and highly speeds up the process of TMA construction and provides a possibility for fully automated IHC quantification.

Reasoning about orchestrations of web services using partial correctness

A service is a remote computational facility which is made available for general use by means of a wide-area network. Several types of service arise in practice: stateless services, shared state services and services with states which are customised for individual users. A service-based orchestration is a multi-threaded computation which invokes remote services in order to deliver results back to a user (publication). In this paper a means of specifying services and reasoning about the correctness of orchestrations over stateless services is presented. As web services are potentially unreliable the termination of even finite orchestrations cannot be guaranteed. For this reason a partial-correctness powerdomain approach is proposed to capture the semantics of recursive orchestrations.

Additivity Pretraining and cue competition effects: Developmental evidence for a reasoning account of causal learning

The effect of additivity pretraining on blocking has been taken as evidence for a reasoning account of human and animal causal learning. If inferential reasoning underpins this effect, then developmental differences in the magnitude of this effect in children would be expected. Experiment 1 examined cue competition effects in children's (4- to 5-year-olds and 6- to 7-year-olds) causal learning using a new paradigm analogous to the food allergy task used in studies of human adult causal learning. Blocking was stronger in the older than the younger children, and additivity pretraining only affected blocking in the older group. Unovershadowing was not affected by age or by pretraining. In experiment 2, levels of blocking were found to be correlated with the ability to answer questions that required children to reason about additivity. Our results support an inferential reasoning explanation of cue competition effects. (c) 2012 APA, all rights reserved.

Children’s Reasoning About the Temporal Order of Past and Future Events

Four- and five-year-olds completed two sets of tasks that involved reasoning about the temporal order in which events had occurred in the past or were to occur in the future. Four-year-olds succeeded on the tasks that involved reasoning about the order of past events but not those that involved reasoning about the order of future events, whereas 5-year-olds passed both types of tasks. Individual children who failed the past-event tasks were not particularly likely to fail the more difficult future-event tasks. However, children's performance on the reasoning tasks was predictive of their performance on a task assessing their comprehension of the terms “before” and “after.” Our results suggest that there may be a developmental change over this age range in the ability to flexibly represent and reason about the before-and-after relationships between events.

A New Model for Automated Examination Timetabling

Automated examination timetabling has been addressed by a wide variety of methodologies and techniques over the last ten years or so. Many of the methods in this broad range of approaches have been evaluated on a collection of benchmark instances provided at the University of Toronto in 1996. Whilst the existence of these datasets has provided an invaluable resource for research into examination timetabling, the instances have significant limitations in terms of their relevance to real-world examination timetabling in modern universities. This paper presents a detailed model which draws upon experiences of implementing examination timetabling systems in universities in Europe, Australasia and America. This model represents the problem that was presented in the 2nd International Timetabling Competition (ITC2007). In presenting this detailed new model, this paper describes the examination timetabling track introduced as part of the competition. In addition to the model, the datasets used in the competition are also based on current real-world instances introduced by EventMAP Limited. It is hoped that the interest generated as part of the competition will lead to the development, investigation and application of a host of novel and exciting techniques to address this important real-world search domain. Moreover, the motivating goal of this paper is to close the currently existing gap between theory and practice in examination timetabling by presenting the research community with a rigorous model which represents the complexity of the real-world situation. In this paper we describe the model and its motivations, followed by a full formal definition.