Guiding local regression using visualisation

Solving many scientific problems requires effective regression and/or classification models for large high-dimensional datasets. Experts from these problem domains (e.g. biologists, chemists, financial analysts) have insights into the domain which can be helpful in developing powerful models but they need a modelling framework that helps them to use these insights. Data visualisation is an effective technique for presenting data and requiring feedback from the experts. A single global regression model can rarely capture the full behavioural variability of a huge multi-dimensional dataset. Instead, local regression models, each focused on a separate area of input space, often work better since the behaviour of different areas may vary. Classical local models such as Mixture of Experts segment the input space automatically, which is not always effective and it also lacks involvement of the domain experts to guide a meaningful segmentation of the input space. In this paper we addresses this issue by allowing domain experts to interactively segment the input space using data visualisation. The segmentation output obtained is then further used to develop effective local regression models.

GTM-based data visualisation with incomplete data

We analyse how the Generative Topographic Mapping (GTM) can be modified to cope with missing values in the training data. Our approach is based on an Expectation -Maximisation (EM) method which estimates the parameters of the mixture components and at the same time deals with the missing values. We incorporate this algorithm into a hierarchical GTM. We verify the method on a toy data set (using a single GTM) and a realistic data set (using a hierarchical GTM). The results show our algorithm can help to construct informative visualisation plots, even when some of the training points are corrupted with missing values.

Semi-supervised learning of hierarchical latent trait models for data visualisation

An interactive hierarchical Generative Topographic Mapping (HGTM) ¸iteH_GTM has been developed to visualise complex data sets. In this paper, we build a more general visualisation system by extending the HGTM visualisation system in 3 directions: bf (1) We generalize HGTM to noise models from the exponential family of distributions. The basic building block is the Latent Trait Model (LTM) developed in ¸iteKaban_pami. bf (2) We give the user a choice of initializing the child plots of the current plot in either em interactive, or em automatic mode. In the interactive mode the user interactively selects ``regions of interest'' as in ¸iteH_GTM, whereas in the automatic mode an unsupervised minimum message length (MML)-driven construction of a mixture of LTMs is employed. bf (3) We derive general formulas for magnification factors in latent trait models. Magnification factors are a useful tool to improve our understanding of the visualisation plots, since they can highlight the boundaries between data clusters. The unsupervised construction is particularly useful when high-level plots are covered with dense clusters of highly overlapping data projections, making it difficult to use the interactive mode. Such a situation often arises when visualizing large data sets. We illustrate our approach on a toy example and apply our system to three more complex real data sets.

Data visualisation with missing data: A non-linear approach

Exploratory analysis of data in all sciences seeks to find common patterns to gain insights into the structure and distribution of the data. Typically visualisation methods like principal components analysis are used but these methods are not easily able to deal with missing data nor can they capture non-linear structure in the data. One approach to discovering complex, non-linear structure in the data is through the use of linked plots, or brushing, while ignoring the missing data. In this technical report we discuss a complementary approach based on a non-linear probabilistic model. The generative topographic mapping enables the visualisation of the effects of very many variables on a single plot, which is able to incorporate far more structure than a two dimensional principal components plot could, and deal at the same time with missing data. We show that using the generative topographic mapping provides us with an optimal method to explore the data while being able to replace missing values in a dataset, particularly where a large proportion of the data is missing.

Block GTM: Incorporating prior knowledge of covariance structure in data visualisation

Visualising data for exploratory analysis is a big challenge in scientific and engineering domains where there is a need to gain insight into the structure and distribution of the data. Typically, visualisation methods like principal component analysis and multi-dimensional scaling are used, but it is difficult to incorporate prior knowledge about structure of the data into the analysis. In this technical report we discuss a complementary approach based on an extension of a well known non-linear probabilistic model, the Generative Topographic Mapping. We show that by including prior information of the covariance structure into the model, we are able to improve both the data visualisation and the model fit.

The visualisation of vitreous using surface modified poly(lactic-co-glycolic acid) microparticles

AIMS To demonstrate the potential use of in vitro poly(lactic-co-glycolic acid) (PLGA) microparticles in comparison with triamcinolone suspension to aid visualisation of vitreous during anterior and posterior vitrectomy. METHODS PLGA microparticles (diameter 10-60 microm) were fabricated using single and/or double emulsion technique(s) and used untreated or following the surface adsorption of a protein (transglutaminase). Particle size, shape, morphology and surface topography were assessed using scanning electron microscopy (SEM) and compared with a standard triamcinolone suspension. The efficacy of these microparticles to enhance visualisation of vitreous against the triamcinolone suspension was assessed using an in vitro set-up exploiting porcine vitreous. RESULTS Unmodified PLGA microparticles failed to adequately adhere to porcine vitreous and were readily washed out by irrigation. In contrast, modified transglutaminase-coated PLGA microparticles demonstrated a significant improvement in adhesiveness and were comparable to a triamcinolone suspension in their ability to enhance the visualisation of vitreous. This adhesive behaviour also demonstrated selectivity by not binding to the corneal endothelium. CONCLUSION The use of transglutaminase-modified biodegradable PLGA microparticles represents a novel method of visualising vitreous and aiding vitrectomy. This method may provide a distinct alternative for the visualisation of vitreous whilst eliminating the pharmacological effects of triamcinolone acetonide suspension.

Dissociating the spatio-temporal characteristics of cortical neuronal activity associated with human volitional swallowing in the healthy adult brain

Human swallowing represents a complex highly coordinated sensorimotor function whose functional neuroanatomy remains incompletely understood. Specifically, previous studies have failed to delineate the temporo-spatial sequence of those cerebral loci active during the differing phases of swallowing. We therefore sought to define the temporal characteristics of cortical activity associated with human swallowing behaviour using a novel application of magnetoencephalography (MEG). In healthy volunteers (n = 8, aged 28-45), 151-channel whole cortex MEG was recorded during the conditions of oral water infusion, volitional wet swallowing (5 ml bolus), tongue thrust or rest. Each condition lasted for 5 s and was repeated 20 times. Synthetic aperture magnetometry (SAM) analysis was performed on each active epoch and compared to rest. Temporal sequencing of brain activations utilised time-frequency wavelet plots of regions selected using virtual electrodes. Following SAM analysis, water infusion preferentially activated the caudolateral sensorimotor cortex, whereas during volitional swallowing and tongue movement, the superior sensorimotor cortex was more strongly active. Time-frequency wavelet analysis indicated that sensory input from the tongue simultaneously activated caudolateral sensorimotor and primary gustatory cortex, which appeared to prime the superior sensory and motor cortical areas, involved in the volitional phase of swallowing. Our data support the existence of a temporal synchrony across the whole cortical swallowing network, with sensory input from the tongue being critical. Thus, the ability to non-invasively image this network, with intra-individual and high temporal resolution, provides new insights into the brain processing of human swallowing. © 2004 Elsevier Inc. All rights reserved.

Spatio-temporal imaging of corticaldesynchronization in migraine visual aura:a magnetoencephalography case study

Objective.-To determine cortical oscillatory changes involved in migraine visual aura using magnetoencephalography (MEG). Background.-Visual aura in the form of scintillating scotoma precedes migraine in many cases. The involvement of cortical spreading depression within striate and extra-striate cortical areas is implicated in the generation of the disturbance, but the details of its progression, the effects on cortical oscillations, and the mechanisms of aura generation are unclear. Methods.-We used MEG to directly image changes in cortical oscillatory power during an episode of scintillating scotoma in a patient who experiences aura without subsequent migraine headache. Using the synthetic aperture magnetometry method of MEG source imaging, focal changes in cortical oscillatory power were observed over a 20-minute period and visualized in coregistration with the patient's magnetic resonance image. Results.-Alpha band desynchronization in both the left extra-striate and temporal cortex persisted for the duration of reported visual disturbance, terminating abruptly upon disappearance of scintillations. Gamma frequency desynchronization in the left temporal lobe continued for 8 to 10 minutes following the reported end of aura. Conclusions.-Observations implicate the extra-striate and temporal cortex in migraine visual aura and suggest involvement of alpha desynchronization in generation of phosphenes and gamma desynchronization in sustained inhibition of visual function.

Advanced data driven visualisation for geo-spatial data

Most current 3D landscape visualisation systems either use bespoke hardware solutions, or offer a limited amount of interaction and detail when used in realtime mode. We are developing a modular, data driven 3D visualisation system that can be readily customised to specific requirements. By utilising the latest software engineering methods and bringing a dynamic data driven approach to geo-spatial data visualisation we will deliver an unparalleled level of customisation in near-photo realistic, realtime 3D landscape visualisation. In this paper we show the system framework and describe how this employs data driven techniques. In particular we discuss how data driven approaches are applied to the spatiotemporal management aspect of the application framework, and describe the advantages these convey.

Grayscale and colour image Codec based on matching pursuit in the spatio-frequency domain

This report presents and evaluates a novel idea for scalable lossy colour image coding with Matching Pursuit (MP) performed in a transform domain. The benefits of the idea of MP performed in the transform domain are analysed in detail. The main contribution of this work is extending MP with wavelets to colour coding and proposing a coding method. We exploit correlations between image subbands after wavelet transformation in RGB colour space. Then, a new and simple quantisation and coding scheme of colour MP decomposition based on Run Length Encoding (RLE), inspired by the idea of coding indexes in relational databases, is applied. As a final coding step arithmetic coding is used assuming uniform distributions of MP atom parameters. The target application is compression at low and medium bit-rates. Coding performance is compared to JPEG 2000 showing the potential to outperform the latter with more sophisticated than uniform data models for arithmetic coder. The results are presented for grayscale and colour coding of 12 standard test images.

Data visualisation and exploration with prior knowledge

Visualising data for exploratory analysis is a major challenge in many applications. Visualisation allows scientists to gain insight into the structure and distribution of the data, for example finding common patterns and relationships between samples as well as variables. Typically, visualisation methods like principal component analysis and multi-dimensional scaling are employed. These methods are favoured because of their simplicity, but they cannot cope with missing data and it is difficult to incorporate prior knowledge about properties of the variable space into the analysis; this is particularly important in the high-dimensional, sparse datasets typical in geochemistry. In this paper we show how to utilise a block-structured correlation matrix using a modification of a well known non-linear probabilistic visualisation model, the Generative Topographic Mapping (GTM), which can cope with missing data. The block structure supports direct modelling of strongly correlated variables. We show that including prior structural information it is possible to improve both the data visualisation and the model fit. These benefits are demonstrated on artificial data as well as a real geochemical dataset used for oil exploration, where the proposed modifications improved the missing data imputation results by 3 to 13%.

Exploratory database visualisation: the application and assessment of data and dimensionality reduction

This thesis describes the development of a complete data visualisation system for large tabular databases, such as those commonly found in a business environment. A state-of-the-art 'cyberspace cell' data visualisation technique was investigated and a powerful visualisation system using it was implemented. Although allowing databases to be explored and conclusions drawn, it had several drawbacks, the majority of which were due to the three-dimensional nature of the visualisation. A novel two-dimensional generic visualisation system, known as MADEN, was then developed and implemented, based upon a 2-D matrix of 'density plots'. MADEN allows an entire high-dimensional database to be visualised in one window, while permitting close analysis in 'enlargement' windows. Selections of records can be made and examined, and dependencies between fields can be investigated in detail. MADEN was used as a tool for investigating and assessing many data processing algorithms, firstly data-reducing (clustering) methods, then dimensionality-reducing techniques. These included a new 'directed' form of principal components analysis, several novel applications of artificial neural networks, and discriminant analysis techniques which illustrated how groups within a database can be separated. To illustrate the power of the system, MADEN was used to explore customer databases from two financial institutions, resulting in a number of discoveries which would be of interest to a marketing manager. Finally, the database of results from the 1992 UK Research Assessment Exercise was analysed. Using MADEN allowed both universities and disciplines to be graphically compared, and supplied some startling revelations, including empirical evidence of the 'Oxbridge factor'.

Non-linear hierarchical visualisation

This thesis applies a hierarchical latent trait model system to a large quantity of data. The motivation for it was lack of viable approaches to analyse High Throughput Screening datasets which maybe include thousands of data points with high dimensions. High Throughput Screening (HTS) is an important tool in the pharmaceutical industry for discovering leads which can be optimised and further developed into candidate drugs. Since the development of new robotic technologies, the ability to test the activities of compounds has considerably increased in recent years. Traditional methods, looking at tables and graphical plots for analysing relationships between measured activities and the structure of compounds, have not been feasible when facing a large HTS dataset. Instead, data visualisation provides a method for analysing such large datasets, especially with high dimensions. So far, a few visualisation techniques for drug design have been developed, but most of them just cope with several properties of compounds at one time. We believe that a latent variable model (LTM) with a non-linear mapping from the latent space to the data space is a preferred choice for visualising a complex high-dimensional data set. As a type of latent variable model, the latent trait model can deal with either continuous data or discrete data, which makes it particularly useful in this domain. In addition, with the aid of differential geometry, we can imagine the distribution of data from magnification factor and curvature plots. Rather than obtaining the useful information just from a single plot, a hierarchical LTM arranges a set of LTMs and their corresponding plots in a tree structure. We model the whole data set with a LTM at the top level, which is broken down into clusters at deeper levels of t.he hierarchy. In this manner, the refined visualisation plots can be displayed in deeper levels and sub-clusters may be found. Hierarchy of LTMs is trained using expectation-maximisation (EM) algorithm to maximise its likelihood with respect to the data sample. Training proceeds interactively in a recursive fashion (top-down). The user subjectively identifies interesting regions on the visualisation plot that they would like to model in a greater detail. At each stage of hierarchical LTM construction, the EM algorithm alternates between the E- and M-step. Another problem that can occur when visualising a large data set is that there may be significant overlaps of data clusters. It is very difficult for the user to judge where centres of regions of interest should be put. We address this problem by employing the minimum message length technique, which can help the user to decide the optimal structure of the model. In this thesis we also demonstrate the applicability of the hierarchy of latent trait models in the field of document data mining.

Colour image coding with matching pursuit in the spatio-frequency domain

We present and evaluate a novel idea for scalable lossy colour image coding with Matching Pursuit (MP) performed in a transform domain. The idea is to exploit correlations in RGB colour space between image subbands after wavelet transformation rather than in the spatial domain. We propose a simple quantisation and coding scheme of colour MP decomposition based on Run Length Encoding (RLE) which can achieve comparable performance to JPEG 2000 even though the latter utilises careful data modelling at the coding stage. Thus, the obtained image representation has the potential to outperform JPEG 2000 with a more sophisticated coding algorithm.

A reevaluation of achromatic spatio-temporal vision: nonoriented filters are monocular, they adapt, and can be used for decision making at high flicker speeds

Masking, adaptation, and summation paradigms have been used to investigate the characteristics of early spatio-temporal vision. Each has been taken to provide evidence for (i) oriented and (ii) nonoriented spatial-filtering mechanisms. However, subsequent findings suggest that the evidence for nonoriented mechanisms has been misinterpreted: those experiments might have revealed the characteristics of suppression (eg, gain control), not excitation, or merely the isotropic subunits of the oriented detecting mechanisms. To shed light on this, we used all three paradigms to focus on the ‘high-speed’ corner of spatio-temporal vision (low spatial frequency, high temporal frequency), where cross-oriented achromatic effects are greatest. We used flickering Gabor patches as targets and a 2IFC procedure for monocular, binocular, and dichoptic stimulus presentations. To account for our results, we devised a simple model involving an isotropic monocular filter-stage feeding orientation-tuned binocular filters. Both filter stages are adaptable, and their outputs are available to the decision stage following nonlinear contrast transduction. However, the monocular isotropic filters (i) adapt only to high-speed stimuli—consistent with a magnocellular subcortical substrate—and (ii) benefit decision making only for high-speed stimuli (ie, isotropic monocular outputs are available only for high-speed stimuli). According to this model, the visual processes revealed by masking, adaptation, and summation are related but not identical.