Ebook: Healthgrid Research, Innovation and Business Case

The organisation and the conferences

HealthGrid 2009 (http://berlin2009.healthgrid.org) is the seventh meeting of this open forum for the integration of grid technologies, e-science and e-health methods and their applications in the biomedical and healthcare domains. The principal objective of the HealthGrid conference and of the HealthGrid Association is the exchange and debate of ideas, technologies, solutions and requirements that interest the grid and the life-science communities and are likely to promote the integration of grids into biomedical research and health in the broadest sense. Participation is encouraged for grid middleware and healthgrid application developers, biomedical and health informatics users, ethicists and security experts, and policy makers to participate in a set of multidisciplinary sessions with a common focus on bringing healthgrids closer to real application in the health domain.

HealthGrid conferences have been organized on an annual basis. The first conference, held in 2003 in Lyon (http://lyon2003.healthgrid.org), reflected the need to involve all actors – physicians, scientists and technologists – who might play a role in the application of grid technology to health, whether healthcare or bio-medical research. The second conference, held in Clermont-Ferrand in January 2004 (http://clermont2004.healthgrid.org) reported on the earliest efforts in research, mainly work in progress, from a large number of projects. The third conference in Oxford (http://oxford2005.healthgrid.org) had a major focus on first results and exploration of deployment strategies in healthcare. The fourth conference in Valencia (http://valencia2006.healthgrid.org) aimed at consolidating the collaboration among biologists, healthcare professionals and grid technology experts. The fifth conference in Geneva (http://geneva2007.healthgrid.org) focused on the five domains defined by the European Commission as application areas for ‘vertical integration’ through grids in the biomedical field: molecules, cells, organs, individuals, and populations. For each of these five domains, an invited speaker gave a state of the art address followed by concrete projects. This was a loud signal to the community that the usefulness of grids to potential application domains could be demonstrated at least at the prototype level. This theme was also evident at the sixth conference in Chicago (http://chicago2008.healthgrid.org), which proclaimed its focus as ‘e-Science Meets Biomedical Informatics’. The sixth conference was also a landmark in the history of the organisation HealthGrid – and its newly established affiliate HealthGrid.US – as the first conference to be organised outside Europe. As we put it a year ago, this was a celebration of similarities and differences, a moment to validate models and principles beyond one's familiar shores. The proceedings since the third conference have appeared in the IOS series Studies in Health Technology and Informatics ([1–4]).

While the desire to promote healthgrid applications to real healthcare settings remains central to the community's ambitions, it is also true that the majority of adopters work in academic environments where research is their principal preoccupation. This continues to be reflected in the content of the conference. Nevertheless, we may detect a move in this research towards application in medicine and healthcare, an admittedly slower process than the most optimistic projections, but with worthwhile exemplars to motivate researchers, innovators and those who must take and justify investment decisions.

The deadline for the Chicago conference proceedings more or less coincided with the announcement – I should say, the shock of the announcement – of Jean-Claude Healy's untimely death. I did not know Jean-Claude personally, although he had made a very deep impression on me on the first occasion I met him: the launch in 2002 of the first explicitly health-focussed grid projects to be funded under the Framework Programme 5 of the European Union. He was always a very busy man, as all his formal and informal obituarists have highlighted, so his welcome and introductory remarks had, to someone who had not met him before, a sense of hurry; one could easily imagine that there was also a committee waiting for him to return so that it could continue its deliberations. But on reflection, this was a wrong impression: the urgency arose from his desire to see results from the projects that had just been approved. What now remains with me from that meeting is the passion of his assertion that, while doctors sometimes appear reluctant to embrace new technologies, the potential of our particular technology, healthgrids, was virtually limitless and that the importance of our projects was that they would demonstrate just how much was possible to an often sceptical medical profession. It is salutary to have his memory to remind us, as we endeavour to bring the fruits of our research from the laboratory to the citizen, that there is a real vision of better healthcare through the appropriate and intelligent use of our technologies – and let our success be a tribute to him in the way that would have counted most with him, through the advancement of science and healthcare. Indeed, it seems entirely fitting that Ilias Iakovidis [5] concludes his intimate tribute in the IMIA Yearbook of Medical Informatics 2008 with the familiar quotation, Jean-Claude's favourite, from George Bernard Shaw about the reasonable and the unreasonable man.

Berlin 2009

The themes of the conference reflected the anticipated, though slow, move towards real applications. Papers were invited in, though not limited to, certain prominent areas and topics. First among them, perhaps for obvious reasons, Accessibility, the challenge of making Grid more accessible to bio-medical users and the fraught issue of usability: “ready-to-run” healthgrids? The next two focal areas were well known to be active among researchers: one was Core Technologies and Data Integration, with the contrast of grid technology versus web applications, data privacy: confidentiality in distributed medical information systems – and the security challenges, semantic techniques and the challenge of integrating heterogeneous biomedical data, visualization in grids and next generation healthgrids: self adaptive systems. The second was Applications, such as healthgrids for genetics and epidemiological studies, pharmagrids supporting pharma research on grids and, not least, the collaboration area of grid computing and the Virtual Physiological Human. It was also anticipated that the theme of Ethical, Legal, Social and Economic (ELSE) Issues would merit several papers, given the extent to which these were being actively researched. These include topics such as grid business case: sustainability and go-to-market strategies, experiences with production grids in real business, and grid sociology: how to win society over to grids. In the event, and with the impact of road maps that had emerged in the past year joining the two, the other prominent topic, The Future of Grids, merged with ELSE. Indeed, The Future holds some questions that more or less centre on social and economic issues: topics such as ‘cloud’ and on-demand/utility computing and new demands beyond technology.

ACCESSIBILITY

Semantic Security: Specification and Enforcement of Semantic Policies for Security-driven Collaborations, by Sinnott and Doherty, reports on a project on Advanced Grid Authorisation through Semantic Technologies (AGAST) which seeks to address some of the most glaring deficiencies of typical authorisation processes when fine grained access control is necessary. The application is illustrated through projects that have already been reported in HealthGrid conferences and have attracted considerable interest. Mohammad Shahid et al, in A Robust Framework for Rapid Deployment of a Virtual Screening Laboratory, describe a framework which allows complex in silico screening workflows to be rapidly defined and efficiently executed on a grid infrastructure based on Unicore and Meta Scheduling Service. A discussion of results on the Viola/Phosphorus test bed leads to suggestions for further work. David Hoyle et al, in Shared Genomics: High Performance Computing for distributed insights in genomic medical research, consider the problem of genome-wide association studies and the need for high performance computing (HPC) to conduct the necessary statistical analyses and for annotation tools that will be needed to manage the data flows through such analyses. The relative merits of a grid infrastructure vs. clusters rapidly lead to ethical, legal and security concerns. Work towards a ‘Shared Genomics’ User Interface continues. On the other hand, Weisbecker and Falkner, in Service Engineering for Grid Services in Medicine and Life Science, explore the definition and use of services to support various paradigms of cloud or on demand computing, or on demand applications. The work is being conducted in the context of Services@mediGRID and addresses the need for a systematic way to develop services and also the need to comprehend the underlying business model in provisioning application services, especially when service customers may own or need to also be provisioned with infrastructure elements.

CORE TECHNOLOGIES AND DATA INTEGRATION

Tobias Knoch et al present The GLOBE 3D Genome Platform: Towards A Novel System-Biological Paper Tool to Integrate the Huge Complexity of Genome Organization and Function, an exploration of how to tame the three-dimensional genome problem using a system biology approach. This requires the eponymous tool of the title to integrate existing structural, analytic and regulatory views of the genome into a novel architecture. At present this is a ‘paper’ tool which allows a variety of techniques to be applied, but a release of an integrated tool is promised in the near future. In Metadata Extraction using Text Mining, Shivani Seth et al investigate the derivation and use of metadata in e-Science applications which combine multiple services to construct integrated frameworks and novel solutions. Taking numerous successful projects as exemplars, they justify and derive a method for the extraction of metadata from relevant documentation of a given service algorithm, thus collecting the necessary information for the integration of that algorithm into a new workflow. In particular, they examine the process of conversion from the statistical package R to the grid services of GridR. Ignacio Blanquer et al deal with metadata in a very different context: in Using Grid-Enabled Distributed Metadata Database to Index DICOM-SR, they examine the dichotomy between DICOM data (e.g. images) and DICOM-SR, the structured report format of the same family of standards. They specify how to create a DICOM-SR template, how to use standard grid components to share and store DICOM data and how to seek a DICOM-SR tool to validate XML descriptions of structured reports. The value of the work lies in improved efficiency for queries and updates. Further work is designed to lead to a tool that will take integrated data and the metadata catalog and from there provide support for other grid infrastructures. In XTENS – an eXTensible Environment for NeuroScience, Luca Corradi et al offer a new perspective on medical scenarios providing an integrated system to handle queries and to display results in the domain of EEG signals. In the last paper of this section, Ainsworth and Buchan discuss the issue of Preserving consent-for-consent with feasibility-assessment and recruitment in clinical studies: FARSITE architecture, to deal with the problem of consent to access records to determine whether an individual is eligible to participate in a study. It is argued that Feasibility Assessment and Recruitment System for Improving Trial Efficiency, an example e-Lab, provides better recruitment without any compromises and at the same time reduces clinician workloads.

APPLICATIONS

Antje Wolf et al in DockFlow – A Prototypic PharmaGrid for Virtual Screening Integrating Four Different Docking Tools create a prototypic pharmagrid, i.e. a tool to support pharmaceutical research in silico, identifying molecular matches through ‘docking’ programs and analysing the wide variety of results generated by such methods. In particular, they integrate results from different methods used in various remote centres providing a common platform for their project partners. Matthias Assel et al continue research into ViroLab, reported in previous HealthGrid conferences, in A Collaborative Environment Allowing Clinical Investigations on Integrated Biomedical Databases. Here they create user interfaces to allow the integration of relevant data sources into ViroLab to enable investigation of its use in medical use cases, with investigation of drug effectiveness in the face of viral mutation as a particular example. In a somewhat different vein, Sebastian Canisius et al consider the Application of Grid technology for automated detection of sleep disordered breathing, a major source of ill-health in the west. The application in question identifies such sleep disorders through analysis of standard biosignals; with well known difficulties in the application of as yet unapproved grid technologies in healthcare, the project has been restricted to research and to the analysis of security, accessibility and fault tolerance of their system as a basis for further progress. In another of many MediGRID projects reporting at this conference, Kamen Beronov et al provide a grid-based tool to exploit haemodynamic simulation in Grid computing for detailed hemodynamics-simulation-based planning of endovascular interventions, essentially a virtual vascular surgery environment which allows the computation of blood pressure effects on vessels in different surgical scenarios; they go on to show how this might be used in two specific pathologies, a stenosis and an aneurysm.

Medical imaging has always been a rich source of challenges for healthgrids and the story continues in the same vein here. Frederik Orellana et al in Running Medical Image Analysis on GridFactory Desktop Grid consider a novel batch system for medical imaging which overcomes some of the traditional barriers in hospital settings by accepting and dealing with the multiple firewall problems that inevitably arise in accessing well protected systems. In the same setting at Geneva Hospitals, Xin Zhou et al in their paper An Easy Setup for Parallel Image-Processing: Using Taverna and ARC present a concrete application which uses well established workflow tools to provide relatively easy parallel processing functionality on a grid. Ralf Lützkendorf et al, on the other hand, in Enabling of Grid based Diffusion Tensor Imaging using a Workflow Implementation of FSL, consider a highly detailed technique for precise MRI imaging of nerve fibres in the brain. This has the merit of allowing in vivo research to be conducted in functional areas of the brain in order to study the relationship between certain kinds of fibre deformation and disease patterns. Silvia Olabarriaga et al are also concerned with MRI methods in brain imaging in Crossing HealthGrid Borders: Early Results in Medical Imaging but the problem they tackle is rather different. Dutch and German grid initiatives both support functional MRI research on both sides of the border, with the scientists wishing to collaborate; enabling the two grid infrastructures to work together provides the occasion of this early experiment in cross-border collaboration. Another organ which is a rich source of imaging problems for healthgrid researchers is the heart. Ketan Maheshwari et al report on their effort to provide a close to healthcare tool in Towards Production-level Cardiac Image Analysis with Grids. They consider two important cardiac image analysis problems, myocardium segmentation and motion estimation, and demonstrate the effectiveness of high level workflow for efficient and robust problem solving.

SOCIO-ECONOMICS AND THE FUTURE OF GRIDS

In studies motivated by the economic question, are healthgrids capable of sustained application in a commercially realistic healthcare environment, two studies report their findings. Alexander Dobrev et al in Economic Performance and Sustainability of HealthGrids: Evidence from Two Case Studies have investigated twenty two exemplars of healthgrids and studied two healthgrid case histories, MammoGrid and WISDOM, in depth and analysed the results. Their succinct conclusion is that “The most significant hurdle to sustainability – the discrepancy between social benefits and private incentives – can be solved by sound business models” – an encouraging message! Indeed, Stefan Scholz et al in Business Aspects and Sustainability for Healthgrids – an Expert Survey report on interviews conducted with 33 experts and concluded that (a) there are some areas of potential economic exploitation of healthgrids, and (b) the expert observations help identify a critical path to the establishment and deployment of real healthcare grids. Also in the economic sphere, Frank Dickmann et al have placed sustainability as their principal criterion of success in Perspectives of MediGRID. They ask, how do grids compare with other emerging paradigms, such as cloud computing, in their application to the life sciences? While at present the grid paradigm is a better fit, the future, as they see it, may require the creation of a ‘stock market’ for high-end computing resources to resolve the forces of supply and demand.

Hanene Rahmouni et al consider Ontology-Based Privacy Compliance on European Healthgrid Domains. The question at the heart of this work is, can high level ethical, legal and regulatory requirements be translated into operational ‘commands’ in a healthgrid environment? A simple example demonstrates the possibilities and the limitations of removing, at least as much as possible, human intervention from the operation of an effective healthgrid infrastructure. One of the ways in which medical knowledge is translated into ‘evidence-based’ practice is through so-called integrated care pathways. In the final paper of this section, Olive and Solomonides consider the potential of Variance analysis as practice-based evidence, i.e. the question how best to interpret variations from the standard pathway, adopted by clinicians in the face of real patients, as opposed to an abstract model, so as to improve the knowledge base: a problem that was first motivated by MammoGrid where the evidence base would have continued to build up from the cases considered in the project.

SHORT PAPERS & POSTERS

As well as the papers presented in the main programme of the conference, a number of short papers arose from poster presentations. These represent an interesting cross-section of imaginative applications of healthgrids.

Raúl Isea et al explore The evolution of HPV by means of a phylogenetic study by exploiting the power of a grid infrastructure and comment on the adequacy of classification systems. Zhuchkov and Tverdokhlebov report on Medical Applications for High-Performance Computers in SKIF-GRID Network, a serious attempt to establish a breast cancer healthgrid in Russia. Paul De Vlieger et al have similarly established a Grid-enabled sentinel network for cancer surveillance, in this case in the Auvergne region where there is considerable support for innovation, by linking cytology, breast cancer screening and epidemiology centres. On the other hand, Eberhard Schmitt et al in Conception of an Image Data Base for Cell Nuclei and Geometric Algorithms for Diagnosis and Therapy Monitoring consider cellular distortions brought about by genome or chromosome changes and examine the diagnostic and staging value of nucleic imaging for certain kinds of cancer.

Stefan Rüping et al, in Workflows for Intelligent Monitoring using Proxy Services, consider some aspects of the problem of translating successful healthgrid research into real healthcare use by up-scaling a proof-of-concept workflow to monitor real data streams. Ashiq Anjum et al report on Reusable Services from the neuGRID Project for Grid-Based Health Applications, a project to encompass research in Alzheimer's disease into the tradition of MammoGrid and Health-e-Child. Gaignard and Montagnat discuss A distributed security policy for neuroradiology data sharing and propose an access control policy to overcome some of the common problems of loss of control over data.

The final short paper by Tobias Knoch et al on e-Human Grid Ecology: Understanding and Approaching the Inverse Tragedy of the Commons in the e-Grid Society is a deliberately provocative invitation to a debate concerning the nature of work ‘in the grid’ and ideas from the movement Scientific Commons.

[1] From Grid to Healthgrid: Proceedings of Healthgrid 2005, T. Solomonides, R. McClatchey, V. Breton, Y. Legré and S. Nørager (Editors), Studies in Health Technology and Informatics Vol 112 IOS Press (2005)

[2] Challenges and Opportunities of Healthgrids: Proceedings of Healthgrid 2006, V. Hernández, I. Blanquer, T. Solomonides, V. Breton and Y. Legré (Editors), Studies in Health Technology and Informatics Vol 120 IOS Press (2006)

[3] From Genes to Personalized HealthCare: Grid Solutions for the Life Sciences: Proceedings of HealthGrid 2007, N. Jacq, Y. Legré, H. Muller, I. Blanquer, V. Breton, D. Hausser, V. Hernández, T. Solomonides and M. Hofmann-Apitius (Editors), Studies in Health Technology and Informatics Vol 126 IOS Press (2007)

[4] Global Healthgrid: e-Science Meets Biomedical Informatics: Proceedings of HealthGrid 2008, T. Solomonides, J.C. Silverstein, J. Saltz, Y. Legré, M. Kratz, I. Foster, V. Breton and J.R. Beck (Editors), Studies in Health Technology and Informatics Vol 138 IOS Press (2008)

[5] A Tribute to Jean-Claude Healy, a Free Thinker and Visionary Leader in Biomedical Informatics, Ilias Iakovidis, IMIA Yearbook of Medical Informatics 2008, Schattauer (2008)

Tony Solomonides

In this paper we present DockFlow, a prototypic version of a PharmaGrid. DockFlow is supporting pharmaceutical research through enabling virtual screening on the Grid. The system was developed in the course of the BRIDGE project funded by the European Commission. Grids have been used before to run compute- and data-intensive virtual screening experiments, like in the WISDOM project. With DockFlow, however, we addressed a variety of problems yet unsolved, like the diversity of results produced by different docking tools. We also addressed the problem of analysing the data produced in a distributed virtual screening system applying a combinatorial docking approach. In DockFlow we worked on a grid-based problem solving environment for virtual screening with the following major features: execution of four different docking services (FlexX, AutoDock, DOCK and GAsDock) at locations in Europe and China remotely from a common workflow, storage of the results in a common Docking Database providing a shared analysis platform for the collaboration partners and combination of the results. The DockFlow prototype is evaluated on two scientific case studies: malaria and avian flu.

At the Geneva University Hospitals work is in progress to establish a computing facility for medical image analysis, potentially using several hundreds of desktop computers. Typically, hospitals do not have a computer infrastructure dedicated to research, nor can the data leave the hospital network for the reasons of privacy. For this purpose, a novel batch system called GridFactory has been tested along-side with the well-known batch system Condor. GridFactory's main benefits, compared to other batch systems, lie in its virtualization support and firewall friendliness.

The tests involved running visual feature extraction from 50 000 anonymized medical images on a small local grid of 20 desktop computers. A comparisons with a Condor based batch system in the same computers is then presented. The performance of GridFactory is found satisfactory.

Sleep related breathing disorders (SRBD) represent a major disease in sleep medicine. For diagnosis and therapy control, extensive overnight investigations are required, encompassing long-term measurement of multiple biosignals in specialized sleep disorders centers. To date, evaluation of the examination is realized by comprehensive visual inspection of the data by an expert. Therefore, many approaches have been made to facilitate diagnosis, among them automated analysis of the ECG signal. In this article, we present a grid based infrastructure for computer aided diagnosis of SRBD, accessible for distributed users. As the analysis algorithms itself are still in a validation phase, and the Grid infrastructure is not approved for clinical applications, the application is currently used for research purposes only. But as important aspects of data-security, accessibility from protected environments, usability and fault-tolerance are already covered, the implementation is a solid base for further enhancement of the platform and paves the way for a sustainable service grid for sleep medicine.

Production exploitation of cardiac image analysis tools is hampered by the lack of proper IT infrastructure in health institutions, the non trivial integration of heterogeneous codes in coherent analysis procedures, and the need to achieve complete automation of these methods. HealthGrids are promising technologies to address these difficulties. This paper details how they can be complemented by high level problem solving environments such as workflow managers to improve the performance of applications both in terms of execution time and robustness of results. Two of the most important important cardiac image analysis tasks are considered, namely myocardium segmentation and motion estimation in a 4D sequence. Results are shown on the corresponding pipelines, using two different execution environments on the EGEE grid production infrastructure.

Medical image processing is known as a computationally expensive and data intensive domain. It is thus well-suited for Grid computing. However, Grid computing usually requires the applications to be designed for parallel processing, which is a challenge for medical imaging researchers in hospitals that are most often not used to this. Making parallel programming methods easier to apply can promote Grid technologies in clinical environments. Readily available, functional tools with an intuitive interface are required to really promote healthgrids. Moreover, the tools need to be well integrated with the Grid infrastructure.

To facilitate the adoption of Grids in the Geneva University Hospitals we have set up a develop environment based on the Taverna workflow engine. Its usage with a medical imaging application on the hospitals' internal Grid cluster is presented in this paper.

In order to perform clinical investigations on integrated biomedical data sets and to predict virological and epidemiological outcome, medical experts require an IT-based collaborative environment that provides them a user-friendly space for building and executing their complex studies and workflows on largely available and high-quality data repositories. In this paper, the authors introduce such a novel collaborative working environment a so-called virtual laboratory for clinicians and medical researchers, which allows users to interactively access and browse several biomedical research databases and re-use relevant data sets within own designed experiments. Firstly, technical details on the integration of relevant data resources into the virtual laboratory infrastructure and specifically developed user interfaces are briefly explained. The second part describes research possibilities for medical scientists including potential application fields and benefits as using the virtual laboratory functionalities for a particular exemplary study.

Image analysis has been strongly present in several healthgrid initiatives from the start, and today we find many imaging projects with successful grid implementations and developments. An example is the analysis of functional MRI data on grids, which has been successfully realized by several projects and that could be of interest for others. However, crossing the borders of existing grids is not trivial because the infrastructures being created for these projects differ, each adopting a (slightly) different software stack. This paper describes our early attempts to cross the borders between the German and Dutch grid infrastructures for medical imaging, motivated by a true wish to share expertise about fMRI analysis on grids between these two communities. We describe how we used off-the-shelf, production-level, grid technology to implement supporting mechanisms for cooperation in fMRI at several levels (users, data, software, workflows and computing resources). This simple exercise provided us valuable insights into the problems of crossing the borders of real grids from a user's perspective. Besides technical aspects, we observed that security and usability are very important for the success of inter-operation of Healthgrid.

Tensor analysis of diffusion weighted magnetic resonance images is increasingly used for non-invasive tracking of nerve fibers in the human brain. Diffusion-tensor imaging (DTI) enables in-vivo research on the internal structure of the central nervous system, encompassing interconnection of functional areas, correlation between fiber deformations and certain desease patterns, as well as brain tumor localization. But the modeling of the local diffusion parameters is a computationally expensive part of the processing pipeline, resulting to run times up to days on standard desktop computers. A grid implementation of the algorithm with slice based parallelization reduces the processing down to 10% compared to a local cluster and 20% compared to sequential processing on the grid. A workflow implementation enables fault-tolerant handling of temporary failures within the grid. Furthermore, pure web-based access to the grid application allows for collaborative utilitzation even from protected infrastructures, as they are typically found in clinical environments.

Detailed numerical simulations of blood flow in arteries with various malformations and its conjugate loads on the vessel walls have been a research topic for specialized medical and engineering communities over decades. The present state of computing resources and software allows access to these elaborate diagnostic and research tools to a broad user circle and even to integrate them into clinical workflows. To tap the full potential of hemodynamic simulations, a Grid-based “virtual vessel surgery” application has been developed and deployed as part of the image processing module of the MediGRID project of the German Federal Ministry of Education and Science (BMBF). The very resource-intensive parts of that application, including not only the numerical simulation itself, but also automated data pre- and post-processing and visualization are implemented and are made available through the MediGRID portal. Some highly interactive workflow steps of the application are left at the user site but automated – via a specially developed Web interface – to a degree allowing intuitive, fast interaction and seamless integration with the Grid components. Standard middleware provided by D-Grid is used throughout. The complete workflow is implemented into the grid and can in principle be carried out using no external software. It was applied to real vessel data with a stenosis and an aneurysm, respectively.

Grid technologies have proven to be very successful in the area of eScience, and healthcare in particular, because they allow to easily combine proven solutions for data querying, integration, and analysis into a secure, scalable framework. In order to integrate the services that implement these solutions into a given Grid architecture, some metadata is required, for example information about the low-level access to these services, security information, and some documentation for the user. In this paper, we investigate how relevant metadata can be extracted from a semi-structured textual documentation of the algorithm that is underlying the service, by the use of text mining methods. In particular, we investigate the semi-automatic conversion of functions of the statistical environment R into Grid services as implemented by the GridR tool by the generation of appropriate metadata.

Genomes are tremendous co-evolutionary holistic systems for molecular storage, processing and fabrication of information. Their system-biological complexity remains, however, still largely mysterious, despite immense sequencing achievements and huge advances in the understanding of the general sequential, three-dimensional and regulatory organization. Here, we present the GLOBE 3D Genome Platform a completely novel grid based virtual “paper” tool and in fact the first system-biological genome browser integrating the holistic complexity of genomes in a single easy comprehensible platform: Based on a detailed study of biophysical and IT requirements, every architectural level from sequence to morphology of one or several genomes can be approached in a real and in a symbolic representation simultaneously and navigated by continuous scale-free zooming within a unique three-dimensional OpenGL and grid driven environment. In principle an unlimited number of multi-dimensional data sets can be visualized, customized in terms of arrangement, shape, colour, and texture etc. as well as accessed and annotated individually or in groups using internal or external data bases/facilities. Any information can be searched and correlated by importing or calculating simple relations in real-time using grid resources. A general correlation and application platform for more complex correlative analysis and a front-end for system-biological simulations both using again the huge capabilities of grid infrastructures is currently under development. Hence, the GLOBE 3D Genome Platform is an example of a grid based approach towards a virtual desktop for genomic work combining the three fundamental distributed resources: i) visual data representation, ii) data access and management, and iii) data analysis and creation. Thus, the GLOBE 3D Genome Platform is the novel system-biology oriented information system urgently needed to access, present, annotate, and to simulate the holistic genome complexity in a unique gateway towards a real understanding, educative presentation and curative manipulation planning of this tremendous evolutionary information grail – genomes.

Integrating medical data at inter-centre level implies many challenges that are being tackled from many disciplines and technologies. Medical informatics have applied an important effort on describing and standardizing Electronic Health Records, and specially standardisation has achieved an important extent on Medical Imaging. Grid technologies have been extensively used to deal with multi-domain authorisation issues and to provide single access points for accessing DICOM Medical Images, enabling the access and processing to large repositories of data. However, this approach introduces the challenge of efficiently organising data according to their relevance and interest, in which the medical report is a key factor. The present work shows an approach to efficiently code radiology reports to enable the multi-centre federation of data resources. This approach follows the tree-like structure of DICOM-SR reports in a self-organising metadata catalogue based on AMGA. This approach enables federating different but compatible distributed repositories, automatically reconfiguring the database structure, and preserving the autonomy of each centre in defining the template. Tools developed so far and some performance results are provided to prove the effectiveness of the approach.

The XTENS (eXTensible Environment for NeuroScience) platform consists in an highly extensible environment for collaborative work that improve repeatability of experiment and provides data storage and analysis capabilities. The platform is divided in repository and application domains, branched in services with different purpose. The first domain is the central component of the platform and consists in a multimodal repository with a client-server architecture. The second one provides remote tools for image and signal visualization and analysis. The main issue for such a platform is not only to provide an extensible collaborative environment, but also to build a development platform for testing models and algorithms in neuroscience. For these reasons a Grid approach has been considered. Both computational and data Grids infrastructures can be exploited to analyze and share large datasets of distributed data. The architecture has been deployed to support surgical planning for patients affected by drug resistant epilepsy. In that scenario, a complex analysis for a fully multimodal dataset including different image modalities, EEG and video is required to localize the origin of the ictal discharge and critical brain areas. As first results, prototype versions of both repository and application domain components are presented.

Best practice guidance for clinical studies asks investigators to employ the highest possible standards in privacy and consent. When considering the feasibility of a clinical study, issues of privacy extend not only to actual but also to potential study participants. The consent required to access records to determine whether or not an individual might be eligible to participate in a study is sometimes referred to as consent-for-consent. Some initiatives to enhance the efficiency of study-recruitment could compromise consent-for-consent, for example by inviting a patient to take part in a study without the knowledge of their attending clinician. Through iterative working with experts and examination of protocols we explored a range of scenarios for assessing the feasibility of clinical trials and observational studies, and recruiting participants. The main requirement we identified was to speed up feasibility-assessment and recruitment while preserving the patient-clinician trust relationship that is central to consent-for-consent. We present an appropriate information system architecture, FARSITE (Feasibility Assessment and Recruitment System for Improving Trial Efficiency), and show in principle that faster recruitment into clinical studies need not compromise best practice in privacy or consent. We show that FARSITE is a specific instance of an ‘e-Lab’ architecture for assembling data, methods and expertise around study protocols and defined populations.

Financial sustainability is not a driving force of HealthGrids today, as a previous desk research survey of 22 international HealthGrid projects has showed. The majority of applications are project based, which puts a time limit of funding, but also of goals and objectives. Given this situation, we analysed two initiatives, WISDOM and MammoGrid from an economic, cost-benefit perspective, and evaluated the potential for these initiatives to be brought to market as self-financing, sustainable services. We conclude that the topic of HealthGrids should be pursued further because of the substantial potential for net gains to society at large. The most significant hurdle to sustainability – the discrepancy between social benefits and private incentives – can be solved by sound business models.

Grid computing initiatives in medicine and life sciences are under pressure to prove their sustainability. While some first business model frameworks were outlined, few practical experiences were considered. This gap has been narrowed by an international survey of 33 grid computing experts with biomedical and non-biomedical background on business aspects. The experts surveyed were cautiously optimistic about a sustainable implementation of grid computing within a mid term timeline. They identified marketable application areas, stated the underlying value proposition, outlined trends and specify critical success factors. From a general perspective of their answers, they provided a stable basis for a road map of sustainable grid computing solutions for medicine and life sciences.

Sustainability is a top priority for nearly all grid communities. The German grid communities in the area of life sciences are continuing their dissemination efforts in order to bring the grid to scientists. With cloud computing another concept for distributed IT infrastructures is on the rise. In this regard the grid has a different focus and matches better with life science compute power demands. A comparison of both grid and cloud in addition to the background and present status of the German life science grid give a contemporary impression of the future perspectives of MediGRID.

The harmonization of data protection law in Europe has been theoretically achieved by means of the EU directive on data protection [1]. In practice the harmonization is not absolute and conflicts continue to exist on the ways member states are implementing the directive. The integration of different European medical systems by means of grid technologies will continue to be challenging if technology does not intervene to enhance interoperability between national regulatory frameworks on data protection. In this paper we present an approach to automate privacy requirements for the sharing of patient data across Europe on a healthgrid [2] domain and ensure its enforcement internally and within external domains where the data might travel. This approach is based on the semantic modelling of privacy obligations that are of legal, ethical or cultural nature. These requirements are for the sharing of personal data between different European member states. Our model reflects both similarities and conflicts, if any, between the different member states. This will allow us to reason on the safeguards a data controller should ask from an organization belonging to another member state before disclosing medical data to them. The system will also generate the relevant set of policies to be enforced at the process level of the grid to ensure privacy compliance before allowing access to the data.

Integrated care pathways, a fine-grained form of medical guideline including the explicit recording of any deviation, or ‘variance’, have been perceived by some as overly prescriptive, limiting clinical freedom and promoting ‘cookbook medicine’. However, feeding the results of the analysis of variance back into the development of a pathway could be an effective way of capturing evidence from practice. This paper summarizes research into the development and use of ICPs, and includes some initial findings from a qualitative study involving clinicians that have helped develop or have used ICPs professionally.

Collaborative research can often have demands on finer-grained security that go beyond the authentication-only paradigm as typified by many e-Infrastructure/Grid based solutions. Supporting finer-grained access control is often essential for domains where the specification and subsequent enforcement of authorization policies is needed. The clinical domain is one area in particular where this is so. However it is the case that existing security authorization solutions are fragile, inflexible and difficult to establish and maintain. As a result they often do not meet the needs of real world collaborations where robustness and flexibility of policy specification and enforcement, and ease of maintenance are essential. In this paper we present results of the JISC funded Advanced Grid Authorisation through Semantic Technologies (AGAST) project (www.nesc.ac.uk/hub/projects/agast) and show how semantic-based approaches to security policy specification and enforcement can address many of the limitations with existing security solutions. These are demonstrated into the clinical trials domain through the MRC funded Virtual Organisations for Trials and Epidemiological Studies (VOTES) project (www.nesc.ac.uk/hub/projects/votes) and the epidemiological domain through the JISC funded SeeGEO project (www.nesc.ac.uk/hub/projects/seegeo).