Investigation of support vector machine for the detection of architectural distortion in mammographic images

This paper investigates detection of architectural distortion in mammographic images using support vector machine. Hausdorff dimension is used to characterise the texture feature of mammographic images. Support vector machine, a learning machine based on statistical learning theory, is trained through supervised learning to detect architectural distortion. Compared to the Radial Basis Function neural networks, SVM produced more accurate classification results in distinguishing architectural distortion abnormality from normal breast parenchyma.

Robustness of mutual information to inter-subject variability for automatic artefact removal from EEG

The externally recorded electroencephalogram (EEG) is contaminated with signals that do not originate from the brain, collectively known as artefacts. Thus, EEG signals must be cleaned prior to any further analysis. In particular, if the EEG is to be used in online applications such as Brain-Computer Interfaces (BCIs) the removal of artefacts must be performed in an automatic manner. This paper investigates the robustness of Mutual Information based features to inter-subject variability for use in an automatic artefact removal system. The system is based on the separation of EEG recordings into independent components using a temporal ICA method, RADICAL, and the utilisation of a Support Vector Machine for classification of the components into EEG and artefact signals. High accuracy and robustness to inter-subject variability is achieved.

Automatic artefact removal from event-related potentials via clustering

This paper outlines a method for automatic artefact removal from multichannel recordings of event-related potentials (ERPs). The proposed method is based on, firstly, separation of the ERP recordings into independent components using the method of temporal decorrelation source separation (TDSEP). Secondly, the novel lagged auto-mutual information clustering (LAMIC) algorithm is used to cluster the estimated components, together with ocular reference signals, into clusters corresponding to cerebral and non-cerebral activity. Thirdly, the components in the cluster which contains the ocular reference signals are discarded. The remaining components are then recombined to reconstruct the clean ERPs.

Small-angle x-ray scattering study of flow alignment of a thermotropic liquid crystal in the nematic and smectic phases

WThe capillary flow alignment of the thermotropic liquid crystal 4-n-octyl-4′-cyanobiphenyl in the nematic and smectic phases is investigated using time-resolved synchrotron small-angle x-ray scattering. Samples were cooled from the isotropic phase to erase prior orientation. Upon cooling through the nematic phase under Poiseuille flow in a circular capillary, a transition from the alignment of mesogens along the flow direction to the alignment of layers along the flow direction (mesogens perpendicular to flow) appears to occur continuously at the cooling rate applied. The transition is centered on a temperature at which the Leslie viscosity coefficient α3 changes sign. The configuration with layers aligned along the flow direction is also observed in the smectic phase. The transition in the nematic phase on cooling has previously been ascribed to an aligning-nonaligning or tumbling transition. At high flow rates there is evidence for tumbling around an average alignment of layers along the flow direction. At lower flow rates this orientation is more clearly defined. The layer alignment is ascribed to surface-induced ordering propagating into the bulk of the capillary, an observation supported by the parallel alignment of layers observed for a static sample at low temperatures in the nematic phase.

Physical interpretation of the correlation between multi-angle spectral data and canopy height

Recent empirical studies have shown that multi-angle spectral data can be useful for predicting canopy height, but the physical reason for this correlation was not understood. We follow the concept of canopy spectral invariants, specifically escape probability, to gain insight into the observed correlation. Airborne Multi-Angle Imaging Spectrometer (AirMISR) and airborne Laser Vegetation Imaging Sensor (LVIS) data acquired during a NASA Terrestrial Ecology Program aircraft campaign underlie our analysis. Two multivariate linear regression models were developed to estimate LVIS height measures from 28 AirMISR multi-angle spectral reflectances and from the spectrally invariant escape probability at 7 AirMISR view angles. Both models achieved nearly the same accuracy, suggesting that canopy spectral invariant theory can explain the observed correlation. We hypothesize that the escape probability is sensitive to the aspect ratio (crown diameter to crown height). The multi-angle spectral data alone therefore may not provide enough information to retrieve canopy height globally.

Surface science studies of strong metal-oxide interactions on model catalysts

The strong metal support interaction (SMSI) was first described in 1978 by Tauster [1-4]. The effect was observed as a severely negative effect on CO and H2 uptake on the catalyst after high temperature calcination under reducing conditions (heating above ~ 700 K) [1,2]. It also had a negative effect on the reaction rate for reactions, such as alkane hydrogenolysis [5,6]. It appeared that the effect occurred for catalysts comprised of reducible supports which were treated at elevated temperature in reducing conditions [2-4]. A classic support which has manifested this behaviour in many studies is TiO2. Over the years following the first discovery of SMSI it has been recognised that the effect is not always negative – for instance for the CO-H2 reaction for which it appears to have a positive effect [5,6]. Further it was noted that hydrogen reduction was not necessary to observe the effect of CO adsorption suppression, it also occurs by vacuum treatment [7], though it should be noted that vacuum treatment at elevated temperature is, in effect, a reducing environment.

Non-stoichiometric oxide and metal interfaces and reactions

We have employed a combination of experimental surface science techniques and density functional calculations to study the reduction of TiO2(110) surfaces through the doping with submonolayer transition metals. We concentrate on the role of Ti adatoms in self doping of rutile and contrast the behaviour to that of Cr. DFT+U calculations enable identification of probable adsorption structures and their spectroscopic characteristics. Adsorption of both metals leads to a broken symmetry and an asymmetric charge transfer localised around the defect site of a mixed localised/delocalised character. Charge transfer creates defect states with Ti 3d character in the band gap at similar to 1-eV binding energy. Cr adsorption, however, leads to a very large shift in the valence-band edge to higher binding energy and the creation of Cr 3d states at 2.8-eV binding energy. Low-temperature oxidation lifts the Ti-derived band-gap states and modifies the intensity of the Cr features, indicative of a change of oxidation state from Cr3+ to Cr4+. Higher temperature processing leads to a loss of Cr from the surface region, indicative of its substitution into the bulk.

Ischaemia differentially regulates GABAB receptor subunits in organotypic hippocampal slice cultures

Reduced synaptic inhibition due to dysfunction of ionotropic GABAA receptors has been proposed as one factor in cerebral ischaemia-induced excitotoxic cell death. However, the participation of the inhibitory metabotropic GABAB receptors in these pathological processes has not been extensively investigated. We used oxygen–glucose deprivation (OGD) and NMDA-induced excitotoxicity as models to investigate whether ischaemia-like challenges alter the protein levels of GABAB1 and GABAB2 receptor subunits in rat organotypic hippocampal slice cultures. Twenty-four hours after the insult both OGD and NMDA produced a marked decrease in the total levels of GABAB2 (75%), while there was no significant change in the levels of GABAB1 after OGD, but an increase after NMDA treatment (100%). The GABAB receptor agonist baclofen (100 μM) was neuroprotective following OGD or NMDA treatment if added before or during the insult. GABAB receptors comprise heterodimers of GABAB1 and GABAB2 subunits and our results suggest that the separate subunits are independently regulated in response to extreme neuronal stress. However, because GABAB2 is required for functional surface expression, down-regulation of this subunit removes an important inhibitory feedback mechanism under pathological conditions.