Multi-class pattern classification in imbalanced data

The majority of multi-class pattern classification techniques are proposed for learning from balanced datasets. However, in several real-world domains, the datasets have imbalanced data distribution, where some classes of data may have few training examples compared for other classes. In this paper we present our research in learning from imbalanced multi-class data and propose a new approach, named Multi-IM, to deal with this problem. Multi-IM derives its fundamentals from the probabilistic relational technique (PRMs-IM), designed for learning from imbalanced relational data for the two-class problem. Multi-IM extends PRMs-IM to a generalized framework for multi-class imbalanced learning for both relational and non-relational domains.

IEF pattern classification-derived criteria for the identification of epoetin-delta in urine

Epoetin-δ (Dynepo™ Shire Pharmaceuticals, Basing stoke, UK) is a synthetic form of erythropoietin (EPO) whose resemblance with endogenous EPO makes it hard to identify using the classical identification criteria. Urine samples collected from six healthy volunteers treated with epoetin-δ injections and from a control population were immuno-purified and analyzed with the usual IEF method. On the basis of the EPO profiles integration, a linear multivariate model was computed for discriminant analysis. For each sample, a pattern classification algorithm returned a bands distribution and intensity score (bands intensity score) saying how representative this sample is of one of the two classes, positive or negative. Effort profiles were also integrated in the model. The method yielded a good sensitivity versus specificity relation and was used to determine the detection window of the molecule following multiple injections. The bands intensity score, which can be generalized to epoetin-α and epoetin-β, is proposed as an alternative criterion and a supplementary evidence for the identification of EPO abuse.

Multi-scale local pattern co-occurrence matrix for textural image classification

Textural image classification technologies have been extensively explored and widely applied in many areas. It is advantageous to combine both the occurrence and spatial distribution of local patterns to describe a texture. However, most existing state-of-the-art approaches for textural image classification only employ the occurrence histogram of local patterns to describe textures, without considering their co-occurrence information. And they are usually very time-consuming because of the vector quantization involved. Moreover, those feature extraction paradigms are implemented at a single scale. In this paper we propose a novel multi-scale local pattern co-occurrence matrix (MS_LPCM) descriptor to characterize textural images through four major steps. Firstly, Gaussian filtering pyramid preprocessing is employed to obtain multi-scale images; secondly, a local binary pattern (LBP) operator is applied on each textural image to create a LBP image; thirdly, the gray-level co-occurrence matrix (GLCM) is utilized to extract local pattern co-occurrence matrix (LPCM) from LBP images as the features; finally, all LPCM features from the same textural image at different scales are concatenated as the final feature vectors for classification. The experimental results on three benchmark databases in this study have shown a higher classification accuracy and lower computing cost as compared with other state-of-the-art algorithms.

Distribution of Gaba in the nerve ganglia of haliotis asinina linnaeus

Gamma-aminobutyric acid (GABA) is a major neurotransmitter and effective settlement inducer in abalone aquaculture. This study aimed to explore the distribution of GABA within neural tissues of Haliotis asinina. Gamma-aminobutyric acid was found in neuronal cell type 1 of 3 major ganglia (i.e., cerebral, pleuropedal, and visceral ganglia) of both sexes. The distribution of GABA-immunoreactive (-ir) cells in the cerebral ganglion was concentrated mostly in the cortex region of the dorsal horn, whereas it was scattered throughout the pleuropedal ganglion, with more in the upper half. Gamma-aminobutyric acid-ir nerve fibers were found throughout the neuropils of the ganglia. The visceral ganglion had the least numbers of GABA-ir neurons compared with the other 2 ganglia. The cells were distributed mainly in the dorsal horn. We also observed GABA to be colocalized with 2 other neurotransmitters: serotonin (5-HT) and dopamine (DA). In the cerebral ganglion, fluorescence double staining of GABA and 5-HT, and GABA and DA showed immunoreactivity in separate cells and was also colocalized in the same cells. In the pleuropedal ganglion, the staining pattern was similar to the cerebral ganglion, but positive-staining cells were less numerous. In the visceral ganglion, GABA and DA, and GABA and 5-HT were colocalized in the same cell types. Overall, we found that GABAergic cells were most numerous in the cerebral ganglion of H. asinina. Further studies are required to determine the functions of these neurotransmitters in relation to their distribution.

Distribution of physical activity facilities in Scotland by small area measures of deprivation and urbanicity

Background The aim of this study was to examine the distribution of physical activity facilities by area-level deprivation in Scotland, adjusting for differences in urbanicity, and exploring differences between and within the four largest Scottish cities. Methods We obtained a list of all recreational physical activity facilities in Scotland. These were mapped and assigned to datazones. Poisson and negative binomial regression models were used to investigate associations between the number of physical activity facilities relative to population size and quintile of area-level deprivation. Results The results showed that prior to adjustment for urbanicity, the density of all facilities lessened with increasing deprivation from quintiles 2 to 5. After adjustment for urbanicity and local authority, the effect of deprivation remained significant but the pattern altered, with datazones in quintile 3 having the highest estimated mean density of facilities. Within-city associations were identified between the number of physical activity facilities and area-level deprivation in Aberdeen and Dundee, but not in Edinburgh or Glasgow. Conclusions In conclusion, area-level deprivation appears to have a significant association with the density of physical activity facilities and although overall no clear pattern was observed, affluent areas had fewer publicly owned facilities than more deprived areas but a greater number of privately owned facilities.

Effects of baffles on flow distribution in an electrostatic precipitator (ESP) of a coal based power plant

Electrostatic Precipitators (ESP) are the most reliable and industrially used control devices to capture fine particles for reducing exhaust emission. Its efficiency is 99% or more. However, capturing submicron particles which are hazardous is still a problem as it involves complex flow phenomena and ESP design limitations. In this study, the effect of baffles on flow distribution inside the ESP is investigated computationally. Baffles are expected to increase the residence time of flue gas which helps to collect more particles into the collector plates, and hence increase the collection efficiency of an ESP. Besides, the placement of a baffle is likely to cause swirling of flue gas and hence sub-micron particles move towards the collector plate due to eccentric and electrostatic force. Therefore, the effects of position, shape and thickness of the baffles on collection efficiency which are also important for ESP design are reported in this study. The fluid flow distribution has been modelled using computational fluid dynamics (CFD) software Fluent and the result and outcome are presented and discussed. The result shows that baffles have significant influence on fluid flow pattern and the efficiency of ESP.

A novel technique for analysing beat-to-beat dynamical changes of QT-RR distribution for arrhythmia prediction

Ventricular tachycardia (VT) leading to ventricular fibrillation (VF) is the major cause of sudden cardiac death (SCD) with subjects with or without any history of cardiac disease. Prediction of the initiation of ventricular fibrillation is crucial for both successful preventive measure and effective defibrillation therapy. A lot of studies have been done based on electrocardiogram (ECG) waveform analysis for VF detection but this field still needs more perfection. Both HRV and QTV related parameters were reported to be analysed for VT/VF detection and prediction with inconsistent results in different populations. In this study, we propose a novel time domain measurement tool to detect the pattern of dynamical changes of both RR and QT intervals in subjects having sustained VT/VF episodes form VFDB and AHA database (www.physionet.org). We also analyse the same pattern in some healthy subjects from Fantasia database and compare the distribution of patterns between healthy and VT/VF subjects. Our findings showed that the distribution of QT-RR dynamics are statistically significantly different (p<0.05) in healthy subjects from VT/VF in particular before the start of VF episode. Therefore, distribution of change in QT-RR dynamics may provide insight of the underlying instability before VF events and can be used for better prediction of arhythmogenesis.