Rule learning and extraction using a hybrid neural network : a case study on fault detection and diagnosis

A hybrid network, based on the integration of Fuzzy ARTMAP (FAM) and the Rectangular Basis Function Network (RecBFN), is proposed for rule learning and extraction problems. The underlying idea for such integration is that FAM operates as a classifier to cluster data samples based on similarity, while the RecBFN acts as a “compressor” to extract and refine knowledge learned by the trained FAM network. The hybrid network is capable of classifying data samples incrementally as well as of acquiring rules directly from data samples for explaining its predictions. To evaluate the effectiveness of the hybrid network, it is applied to a fault detection and diagnosis task by using a set of real sensor data collected from a Circulating Water (CW) system in a power generation plant. The rules extracted from the network are analyzed and discussed, and are found to be in agreement with experts’ opinions used in maintaining the CW system.

Mania : diagnosis and treatment recommendations

This article provides recommendations for the diagnosis and treatment of mania, which characterizes bipolar I disorder (BD I). Failure to detect mania leads to misdiagnosis and suboptimal treatment. To diagnose mania, clinicians should include a detailed mood history within their assessment of patients presenting with depression, agitation, psychosis or insomnia. With regards to treatment, by synthesizing the findings from recent treatment guidelines, and reviewing relevant literature, this paper has distilled recommendations for both acute and long-term management. Antimanic agents including atypical antipsychotics and traditional mood stabilizers are employed to reduce acute manic symptoms, augmented by benzodiazepines if needed, and in refractory or severe cases with behavioural and/or psychotic disturbance, electroconvulsive therapy may occasionally be necessary. Maintenance/prophylaxis therapy aims to reduce recurrences/relapse, for which the combination of psychological interventions with pharmacotherapy is beneficial as it ensures adherence and monitoring of tolerability.

Early diagnosis and treatment of necrotizing fasciitis can improve survival : an observational intensive care unit cohort study

Background
The aim of this study was to describe the clinical characteristics, causative pathogens, clinical management and outcomes of patients presenting to a tertiary adult Australian intensive care unit (ICU) with a diagnosis of necrotizing fasciitis (NF).
Methods
This retrospective observational study was conducted in a 19-bed, level III, adult ICU in a 450-bed tertiary, regional hospital. Clinical databases were accessed for patients diagnosed with NF and admitted to The Geelong Hospital ICU between 1 February 2000 and 1 June 2011. Information on severity of sepsis, surgical procedures and microbiological results were collected.
Results
Twenty patients with NF were identified. The median age was 52.5 years and 38% were female. The overall mortality rate was 8.3%. Common co-morbidities were diabetes (21%) and heart failure (17%), although 50% of patients had no co-morbidities. Group A Streptococcus was the identified pathogen in 11 (46%) patients, and Streptococcus milleri group in 5 (21%) patients. Hyperbaric oxygen therapy was not used in the majority of patients. The initial antibiotics administered were active against subsequently cultured bacteria in 83% of patients. Median time to surgical debridement was 20 h. Diagnosis and management was delayed in the nosocomial group.
Conclusions
This study reports physiological data, aetiology and therapeutic interventions in NF for an adult tertiary hospital. We demonstrate one of the lowest reported mortality rates, with early surgical debridement being achieved in the majority of patients. The main delay was found to be in the diagnosis of NF.

Application of the fuzzy min-max neural network to fault detection and diagnosis of induction motors

In this paper, an application of the motor current signature analysis (MCSA) method and the fuzzy min–max (FMM) neural network to detection and classification of induction motor faults is described. The finite element method is employed to generate simulated data pertaining to changes in the stator current signatures under different motor conditions. The MCSA method is then used to process the stator current signatures. Specifically, the power spectral density is employed to extract harmonics features for fault detection and classification with the FMM network. Various types of induction motor faults, which include stator winding faults and eccentricity problems, under different load conditions are experimented. The results are analyzed and compared with those from other methods. The outcomes indicate that the proposed technique is effective for fault detection and diagnosis of induction motors under different conditions.

Is diagnosis enough to guide interventions in mental health? Using case formulation in clinical practice

While diagnosis has traditionally been viewed as an essential concept in medicine, particularly when selecting treatments, we suggest that the use of diagnosis alone may be limited, particularly within mental health. The concept of clinical case formulation advocates for collaboratively working with patients to identify idiosyncratic aspects of their presentation and select interventions on this basis. Identifying individualized contributing factors, and how these could influence the person's presentation, in addition to attending to personal strengths, may allow the clinician a deeper understanding of a patient, result in a more personalized treatment approach, and potentially provide a better clinical outcome.

Diagnosis of myalgic encephalomyelitis : where are we now?

Introduction: The World Health Organization has classified myalgic encephalomyelitis (ME) as a neurological disease since 1969 considering chronic fatigue syndrome (CFS) as a synonym used interchangeably for ME since 1969. ME and CFS are considered to be neuro-immune disorders, characterized by specific symptom profiles and a neuro-immune pathophysiology. However, there is controversy as to which criteria should be used to classify patients with “chronic fatigue syndrome.”

Areas covered: The Centers for Disease Control and Prevention (CDC) criteria consider chronic fatigue (CF) to be distinctive for CFS, whereas the International Consensus Criteria (ICC) stresses the presence of post-exertion malaise (PEM) as the hallmark feature of ME. These case definitions have not been subjected to rigorous external validation methods, for example, pattern recognition analyses, instead being based on clinical insights and consensus.

Expert opinion: Pattern recognition methods showed the existence of three qualitatively different categories: (a) CF, where CF evident, but not satisfying full CDC syndrome criteria. (b) CFS, satisfying CDC criteria but without PEM. (c) ME, where PEM is evident in CFS. Future research on this “chronic fatigue spectrum” should, therefore, use the abovementioned validated categories and novel tailored algorithms to classify patients into ME, CFS, or CF.

The use of sensitive chemical antibodies for diagnosis: detection of low levels of Epcam in breast cancer

EpCAM is expressed at low levels in a variety of normal human epithelial tissues, but is overexpressed in 70–90% of carcinomas. From a clinico-pathological point of view, this has both prognostic and therapeutic significance. EpCAM was first suggested as a therapeutic target for the treatment of epithelial cancers in the 1990s. However, following several immunotherapy trials, the results have been mixed. It has been suggested that this is due, at least in part, to an unknown level of EpCAM expression in the tumors being targeted. Thus, selection of patients who would benefit from EpCAM immunotherapy by determining EpCAM status in the tumor biopsies is currently undergoing vigorous evaluation. However, current EpCAM antibodies are not robust enough to be able to detect EpCAM expression in all pathological tissues.

Here we report a newly developed EpCAM RNA aptamer, also known as a chemical antibody, which is not only specific but also more sensitive than current antibodies for the detection of EpCAM in formalin-fixed paraffin-embedded primary breast cancers. This new aptamer, together with our previously described aptamer, showed no non- specific staining or cross-reactivity with tissues that do not express EpCAM. They were able to reliably detect target proteins in breast cancer xenograft where an anti-EpCAM antibody (323/A3) showed limited or no reactivity. Our results demonstrated a more robust detection of EpCAM using RNA aptamers over antibodies in clinical samples with chromogenic staining. This shows the potential of aptamers in the future of histopathological diagnosis and as a tool to guide targeted immunotherapy.