Transformers have rapidly become the dominant architecture for analyzing sequential data, utilizing their self-attention mechanism to effectively capture long-term temporal patterns, outperforming recurrent-based methods across various applications. In this paper, we explore the application of transformers to wearable sensor data, focusing on the analysis of human gait, which is often complex and sensitive. We propose two novel frameworks: Data Efficient Sensor Transformer (DesT) for centralized learning and Federated Data Efficient Sensor Transformer (FeDesT) for federated learning (FL) in edge-computing environments. Both frameworks employ knowledge distillation to improve the generalization of transformers, which can be prone to over-fitting due to the limited labeled data available in wearable sensor applications. Experimental results using human gait data collected from uneven and irregular surfaces show that DesT improves the accuracy by 14.8% when compared to existing transformers. FeDesT reduces computational demands on edge devices while outperforming traditional FL methods for transformers. This work demonstrates the potential of transformers for wearable sensor data analysis in both centralized and federated contexts, particularly where privacy and computational efficiency is paramount.

This poster describes the data analysis infrastructure used in a large EU medical project, describes our experiences and lessons learned and presents our techniques for improving the reliability of the analysis procedure.

Presentation for IEEE e-Science 20202

Gait data collected using wearable sensors offers non-intrusive, affordable, real-time monitoring of human motion. Recognizing surface conditions from wearable sensor data has the potential to help systems discriminate between ‘poor quality’ walking data. This research investigates the predictive capabilities of machine learning models, trained on both centralized and decentralized datasets, at categorizing uneven and irregular surface conditions. The results showed that machine learning classification algorithms, trained with data originating from a single sensor positioned on the left-shank, were able to accurately discriminate between different types of surface conditions. We found the Support Vector Machine, when trained with the data centralized, had a test-set accuracy of 94%. Federated Learning offers a way to increase privacy and security for healthcare applications by avoiding the centralization of data. Our simulated federated Deep Neural Network converged to a test-accuracy of 85%, which was 8% less than the centralized counterpart.

There are many scenarios in which it is necessary to collect data from multiple sources in order to evaluate a system, including the collection of both quantitative data - from sensors and smart devices - and qualitative data - such as observations and interview results. However, there are currently very few systems that enable both of these data types to be combined in such a way that they can be analysed side-by-side. This paper describes an end-to-end system for the collection, analysis, storage and visualisation of qualitative and quantitative data, developed using the e-Science Central cloud analytics platform. We describe the experience of developing the system, based on a case study that involved collecting data about the built environment and its users. In this case study, data is collected from older adults living in residential care. Sensors were placed throughout the care home and smart devices were issued to the residents. This sensor data is uploaded to the analytics platform and the processed results are stored in a data warehouse, where it is integrated with qualitative data collected by healthcare and architecture researchers. Visualisations are also presented which were intended to allow the data to be explored and for potential correlations between the quantitative and qualitative data to be investigated.

The use of cloud resources for processing and analysing medical data has the potential to revolutionise the treatment of a number of chronic conditions. For example, it has been shown that it is possible to manage conditions such as diabetes, obesity and cardiovascular disease by increasing the right forms of physical activity for the patient. Typically, movement data is collected for a patient over a period of several weeks using a wrist worn accelerometer. This data, however, is large and its analysis can require significant computational resources. Cloud computing offers a convenient solution as it can be paid for as needed and is capable of scaling to store and process large numbers of data sets simultaneously. However, because the charging model for the cloud represents, to some extent, an unknown cost and therefore risk to project managers, it is important to have an estimate of the likely data processing and storage costs that will be required to analyse a set of data. This could take the form of data collected from a patient in clinic or of entire cohorts of data collected from large studies. If, however, an accurate model was available that could predict the compute and storage requirements associated with a piece of analysis code, decisions could be made as to the scale of resources required in order to obtain results within a known timescale. This paper makes use of provenance and performance data collected as part of routine e-Science Central workflow executions to examine the feasibility of automatically generating predictive models for workflow execution times based solely on observed characteristics such as data volumes processed, algorithm settings and execution durations. The utility of this approach will be demonstrated via a set of benchmarking examples before being used to model workflow executions performed as part of two large medical movement analysis studies.

Summary One of the foundations of science is that researchers must publish the methodology used to achieve their results so that others can attempt to reproduce them. This has the added benefit of allowing methods to be adopted and adapted for other purposes. In the field of e‐Science, services – often choreographed through workflow, process data to generate results. The reproduction of results is often not straightforward as the computational objects may not be made available or may have been updated since the results were generated. For example, services are often updated to fix bugs or improve algorithms. This paper addresses these problems in three ways. Firstly, it introduces a new framework to clarify the range of meanings of ‘reproducibility’. Secondly, it describes a new algorithm, PDIFF, that uses a comparison of workflow provenance traces to determine whether an experiment has been reproduced; the main innovation is that if this is not the case then the specific point(s) of divergence are identified through graph analysis, assisting any researcher wishing to understand those differences. One key feature is support for user‐defined, semantic data comparison operators. Finally, the paper describes an implementation of PDIFF that leverages the power of the e‐Science Central platform that enacts workflows in the cloud. As well as automatically generating a provenance trace for consumption by PDIFF, the platform supports the storage and reuse of old versions of workflows, data and services; the paper shows how this can be powerfully exploited to achieve reproduction and reuse. Copyright © 2013 John Wiley & Sons, Ltd.

Amongst the therapies available to stroke sufferers, one that is gaining attention is the application of video games to encourage therapeutic movement. The Limbs Alive project at Newcastle University has developed a system that gathers therapeutic game data from patients, uses statistical tools to estimate a number of performance metrics and presents the results to patients and clinicians via web applications. This paper describes the architecture of this system and outlines the various technical challenges that were overcome, including in security and deployment.

Abstract Scientists face many severe challenges in extracting value from the increasingly large volumes of data they generate. In this paper we describe the requirements we have derived from working across a wide range of e‐science projects. In particular, the CARMEN neuroinformatics project has exposed a range of challenges due to a need to analyse and share large volumes of data. We have identified the four key activities required by scientists with whom we work, and designed an integrated system—e‐Science Central—to provide them. This exploits three emerging technologies: software as a service to avoid the need for users to deploy and maintain any of their own software; social networking to allow users to collaborate by sharing data, services and workflows in a controlled manner and Cloud computing to provide scalable compute resources. The system can not only be used through any web browser, but also provides an API so that applications can build on the core functionality. We describe the requirements, and the design that flows from them. This includes data storage with in‐built versioning and signing, an in‐browser workflow editor and a job scheduling system that allows workflows to be run both on local ‘private’ clouds and the Microsoft Azure Cloud. Copyright © 2010 John Wiley & Sons, Ltd.

The need for computational resources capable of processing geospatial data has accelerated the uptake of geospatial web services. Several academic and commercial organizations now offer geospatial web services for data provision, coordinate transformation, geocoding and several other tasks. These web services adopt specifications developed by the Open Geospatial Consortium (OGC) - the leading standardization body for Geographic Information Systems. In parallel with efforts of the OGC, the Grid computing community has published specifications for developing Grid applications. The Open Grid Forum (OGF) is the main body that promotes interoperability between Grid computing systems. This study examines the integration of Grid services and geospatial web services into workflows for Geoscientific processing. An architecture is proposed that bridges web services based on the abstract geospatial architecture (ISO19119) and the Open Grid Services Architecture (OGSA). The paper presents a workflow management system, called SAW-GEO, that supports orchestration of Grid-enabled geospatial web services. An implementation of SAW-GEO is presented, based on both the Simple Conceptual Unified Flow Language (SCUFL) and the Business Process Execution Language for Web Services (WS-BPEL or BPEL for short).

Abstract The paper discusses the GOLD project (Grid‐based Information Models to Support the Rapid Innovation of New High Value‐Added Chemicals) whose principal aim is to carry out research and development into enabling technologies to support the formation, operation and termination of virtual organizations. The paper discusses the outcome of this research, which is the GOLD Middleware infrastructure. The infrastructure has been implemented in the form of a set of Middleware components, which address issues such as trust, security, contract monitoring and enforcement, information management and coordination. We discuss all these issues in turn and more importantly we demonstrate how current WS standards can be used to implement these issues. In addition, the paper follows a top down approach starting with a brief outline on the architectural elements derived during the requirements engineering phase and demonstrates how these elements were mapped onto actual services that were implemented according to service‐oriented architecture principles and related technologies. Copyright © 2008 John Wiley & Sons, Ltd.