Uncertainty and topographic visualisations

This thesis is a study of low-dimensional visualisation methods for data visualisation under certainty of the input data. It focuses on the two main feed-forward neural network algorithms which are NeuroScale and Generative Topographic Mapping (GTM) by trying to make both algorithms able to accommodate the uncertainty. The two models are shown not to work well under high levels of noise within the data and need to be modified. The modification of both models, NeuroScale and GTM, are verified by using synthetic data to show their ability to accommodate the noise. The thesis is interested in the controversy surrounding the non-uniqueness of predictive gene lists (PGL) of predicting prognosis outcome of breast cancer patients as available in DNA microarray experiments. Many of these studies have ignored the uncertainty issue resulting in random correlations of sparse model selection in high dimensional spaces. The visualisation techniques are used to confirm that the patients involved in such medical studies are intrinsically unclassifiable on the basis of provided PGL evidence. This additional category of ‘unclassifiable’ should be accommodated within medical decision support systems if serious errors and unnecessary adjuvant therapy are to be avoided.

Predictive gene lists for breast cancer prognosis: a topographic visualisation study

Background: The controversy surrounding the non-uniqueness of predictive gene lists (PGL) of small selected subsets of genes from very large potential candidates as available in DNA microarray experiments is now widely acknowledged 1. Many of these studies have focused on constructing discriminative semi-parametric models and as such are also subject to the issue of random correlations of sparse model selection in high dimensional spaces. In this work we outline a different approach based around an unsupervised patient-specific nonlinear topographic projection in predictive gene lists. Methods: We construct nonlinear topographic projection maps based on inter-patient gene-list relative dissimilarities. The Neuroscale, the Stochastic Neighbor Embedding(SNE) and the Locally Linear Embedding(LLE) techniques have been used to construct two-dimensional projective visualisation plots of 70 dimensional PGLs per patient, classifiers are also constructed to identify the prognosis indicator of each patient using the resulting projections from those visualisation techniques and investigate whether a-posteriori two prognosis groups are separable on the evidence of the gene lists. A literature-proposed predictive gene list for breast cancer is benchmarked against a separate gene list using the above methods. Generalisation ability is investigated by using the mapping capability of Neuroscale to visualise the follow-up study, but based on the projections derived from the original dataset. Results: The results indicate that small subsets of patient-specific PGLs have insufficient prognostic dissimilarity to permit a distinction between two prognosis patients. Uncertainty and diversity across multiple gene expressions prevents unambiguous or even confident patient grouping. Comparative projections across different PGLs provide similar results. Conclusion: The random correlation effect to an arbitrary outcome induced by small subset selection from very high dimensional interrelated gene expression profiles leads to an outcome with associated uncertainty. This continuum and uncertainty precludes any attempts at constructing discriminative classifiers. However a patient's gene expression profile could possibly be used in treatment planning, based on knowledge of other patients' responses. We conclude that many of the patients involved in such medical studies are intrinsically unclassifiable on the basis of provided PGL evidence. This additional category of 'unclassifiable' should be accommodated within medical decision support systems if serious errors and unnecessary adjuvant therapy are to be avoided.

IT as enabler of sustainable farming:an empirical analysis of farmers' adoption decision of precision agriculture technology

Precision agriculture (PA) describes a suite of IT based tools which allow farmers to electronically monitor soil and crop conditions and analyze treatment options. This study tests a model explaining the difficulties of PA technology adoption. The model draws on theories of technology acceptance and diffusion of innovation and is validated using survey data from farms in Canada. Findings highlight the importance of compatibility among PA technology components and the crucial role of farmers' expertise. The model provides the theoretical and empirical basis for developing policies and initiatives to support PA technology adoption.

A sustainable supply chain study of the Indian bioenergy sector

In India, more than one third of the population do not currently have access to modern energy services. Biomass to energy, known as bioenergy, has immense potential for addressing India’s energy poverty. Small scale decentralised bioenergy systems require low investment compared to other renewable technologies and have environmental and social benefits over fossil fuels. Though they have historically been promoted in India through favourable policies, many studies argue that the sector’s potential is underutilised due to sustainable supply chain barriers. Moreover, a significant research gap exists. This research addresses the gap by analysing the potential sustainable supply chain risks of decentralised small scale bioenergy projects. This was achieved through four research objectives, using various research methods along with multiple data collection techniques. Firstly, a conceptual framework was developed to identify and analyse these risks. The framework is founded on existing literature and gathered inputs from practitioners and experts. Following this, sustainability and supply chain issues within the sector were explored. Sustainability issues were collated into 27 objectives, and supply chain issues were categorised according to related processes. Finally, the framework was validated against an actual bioenergy development in Jodhpur, India. Applying the framework to the action research project had some significant impacts upon the project’s design. These include the development of water conservation arrangements, the insertion of auxiliary arrangements, measures to increase upstream supply chain resilience, and the development of a first aid action plan. More widely, the developed framework and identified issues will help practitioners to take necessary precautionary measures and address them quickly and cost effectively. The framework contributes to the bioenergy decision support system literature and the sustainable supply chain management field by incorporating risk analysis and introducing the concept of global and organisational sustainability in supply chains. The sustainability issues identified contribute to existing knowledge through the exploration of a small scale and developing country context. The analysis gives new insights into potential risks affecting the whole bioenergy supply chain.

Support tools for paediatric inpatient prescribers:a review

Prescribing support for paediatrics is diverse and includes both standard texts and electronic tools. Evidence concerning who should be supported and by what method is limited. This review aims to collate the current information available on prescribing support in paediatrics. Many tools designed to support prescribers are technology based. For example, electronic prescribing and smart phone applications. There is a focus on prescriber education both at undergraduate and postgraduate level. In the UK, the majority of inpatient prescribing is done by junior medical staff. It is important to ensure they are competent on qualification and supported in this role. A UK national prescribing assessment is being trialled to test for competence on graduation and there are also tools available to test paediatric prescribing after qualification. No information is available on the tools and resources UK prescribers currently use to support their decision making. One US study reported a decrease in the availability of paediatric prescribing information in a popular reference text. There is limited evidence to show that decisionsupport tools improve patient outcomes, however, there is growing confirmation that electronic prescribing reduces medication errors. There have been reports of new error types, such as selection errors, occurring with the use of electronic prescribing. Another concern with computerised decision-support systems is deciding what alerts should be presented to the prescriber and when/how often in order to avoid alert fatigue. There is little published concerning paediatric alerts perhaps as a consequence of commercial systems often not including paediatric specific support.

Support for paediatric prescribers:what is discussed at hospital board level?

Prescribing support tools range from traditional printed texts to state-of-the-art computerised decision support systems. Comparison between available literature is difficult due to country-specific resources often being the focus of the research. In the UK, it is widely accepted that hospitals take their own individualised approaches to reducing prescribing errors. Objective - This study focused on specialist paediatric hospitals. It aimed to identify the localised approaches taken by paediatric hospitals to reduce prescribing errors. Method - Applied thematic analysis was used to explore the publically published board meeting minutes from the four specialist stand-alone paediatric hospitals in England. Three years of data was collected from each hospital. Codes were collected into groups to identify themes from the data. Results - The main themes identified were clinician involvement in prescribing support is important; credit card-sized reminder tools are used to provide prescribing guidance; electronic prescribing is considered important for reducing prescribing errors; feedback from clinical pharmacists on prescribing errors is widely used; junior doctors require extra support when prescribing; medical records may be incomplete and specific prescribing support (eg, antibiotic prescribing support) is widely in use. Conclusions - There is no single collaborative approach taken to paediatric prescribing support in English paediatric hospitals. Success of electronic prescribing in English paediatric hospitals is considerably behind leaders such as the USA. Use of clinical pharmacists to support prescribers is important as supported by previous studies in Spain and the USA.