Grid Computing, Health Grids, and EHR Systems

Shepherdstown 26 December 2006This article provides a high level overview of grid computing in healthcare at this point in time. In addition, the article also tries to bring into focus the relationship grid computing has to electronic health records (EHR) systems.

Advertisement

Grid Computing

Grid computing is a computing model that provides the ability to perform higher throughput computing by taking advantage of many networked computers to form the equivalent of a virtual super computer. Grid computing utilizes the unused capacity of many separate computers connected by a network to solve large-scale computational problems.

According to Wikipedia, fully functional prototype Grid systems date back to the early 1970s with the Distributed Computing System (DCS) project at the University of California, Irvine. This technology was mostly abandoned in the 1980s as the administrative and security issues involved in having machines you did not control do your computation were seen as too problematic. Many of these issues have since been addressed.

The ideas of the Grid were brought together by Ian Foster, Carl Kesselman and Steve Tuecke, the so-called "fathers of the Grid". They led the effort to create the Globus Toolkit incorporating CPU management, storage management, security, data exchange, and tools for developing additional services such as notification mechanisms, trigger services and information aggregation. The open source Globus Toolkit remains the de facto standard for building Grid solutions, although a number of other noteworthy tools have also been built. See http://en.wikipedia.org/wiki/Grid_computing

At its core, grid computing is based on an open set of standards and protocols - e.g., Open Grid Services Architecture (OGSA) - that enable communication across heterogeneous, geographically dispersed environments. With grid computing, organizations can optimize both internal and external computing and data resources, pool them for large capacity workloads, share them across networks and enable collaboration. See http://www-1.ibm.com/grid/about_grid/what_is.shtml

"Grid" Organizations

Some of the key organizations involved with grid solutions and activities include:

"Grid" Architecture, Standards & Tools

Some of the key web sites to visit when seeking information about grid architecture, standards, and tools include:

"Grid" Publications

The following are selected publications focused exclusively on grids that one should consider looking over:

Examples of Operational Grids

Substantial experience is being built through the operation of various types of grids that have been deployed around the world by both public and private sector organizations. Examples of selected grids to look at include:

Grids in Healthcare

Grids are being used in healthcare in a number of ways. There are now at least three types of grids that have emerged. There are: (1) computational grids being used to solve large-scale computation problems in healthcare research; (2) data grids that don't share computing power but instead provide a standardized way to swap data internally and externally for data mining and decision support; and (3) collaborative grids that let dispersed users share information and work together on extremely large data sets.

The following is a brief list of areas where grid technology is being used in the biology, medical and health related fields:

  • Genetic Linkage Analysis
  • Molecular Sequence Analysis
  • Determination of Protein Structures
  • Identification of Genes and Regulatory Patterns
  • Biological Information Retrieval
  • Biomedical Modeling and Simulation
  • Biomedical Image Processing and Analysis
  • Data Mining and Visualization of Biomedical Data
  • Text Mining of Medical Information Bases

According to a recent article by Leslie Versweyveld, the HealthGrid 2006 proceedings identified medical imaging and bioinformatics as the two main areas in which HealthGrids have demonstrated a greater level of maturity. In the medical imaging field, the NeuroBase Project team is using a federated grid to distribute neuro-imaging information sources across the Internet. Bioinformatics grids are demonstrating their use in the fight against diseases such as malaria in the WISDOM project, infectious diseases in the Sealife project, and SARS with the development of an Access Grid in Taiwan that is serving as a disease management and collaborative platform for medical data exchange. See http://www.hoise.com/vmw/06/articles/vmw/LV-VM-08-06-39.html

The Cancer Biomedical Informatics Grid (caBIG) is an example of a voluntary biomedical informatics network or grid that virtually connects individuals and institutions. It enables both individuals and institutions to effectively share data and research tools in an open environment with common standards. caBIG provides a true world-wide-web of cancer research and development providing a common unification that accelerates understanding of the disease and delivers new approaches for the prevention, early detection, and treatment of cancer. caBIG is being developed under the leadership of the National Cancer Institute's Center for Bioinformatics (NCICB). See https://cabig.nci.nih.gov/

IBM launched a new research effort in 2005 to help battle AIDS using the massive computational power of the World Community Grid, a global community of computer users who joined the philanthropic technology initiative by simply donating unused time on their personal computers. This initiative known as FightAIDS@Home, will deploy massive computer power to develop novel chemical strategies effective in the treatment of HIV-infected individuals in the face of evolving drug resistance in the virus. Developing new, more robust therapies to prevent the onset of AIDS in individuals infected with HIV is the focus of the Olson Laboratory project at The Scripps Research Institute. See http://www.hoise.com/vmw/05/articles/vmw/LV-VM-12-05-23.html

The TeraGrid Bioportal is a shared, extensible portal environment that brings together more than 100 applications and many standard biological data sets. It provides access to high-end computing resources, including a dedicated cluster and TeraGrid systems across the United States. Through the Bioportal, biological researchers, students and educators are able to seamlessly access data, resources and applications, compare biological data stored in different formats and remotely collaborate with colleagues. The portal builds on the success of the North Carolina Bioportal used by researchers and educators in North Carolina. See http://www.hoise.com/vmw/06/articles/vmw/LV-VM-06-06-31.html

The Smallpox Research Grid Project involved screening 35 million potential drug molecules against several protein targets - one of the largest computational chemistry projects ever undertaken. Powering the project was a 2-million device public grid running Grid MP software from United Devices. The Smallpox project is now complete and results have been turned in for post processing. See www.grid.org/projects/smallpox/

The Cancer Research Project is one of several distributed computing projects that have been operated on the Grid.Org web site by United Devices. This project began in 2001 and uses distributed computing power to find drugs for the treatment of cancer. Nearly 300,000 people and their computers participate at this effort, which is also supported by an alliance of organizations such as the National Foundation for Cancer Research and the University of Oxford Department of Chemistry. Visit http://www.grid.org/projects/cancer/

The Grid-Enabled Medical Simulation Services (GEMSS) project is harnessing the grid's processing power to place it in the hands of medical practitioners. The GEMSS project planned to present the first prototype of its Grid middleware at the end of February 2005, along with a test bed that would be one of Europe's first computing and resource grids for clinical use. It would allow easy access to advanced simulation and image processing tools operating at levels of speed and efficiency that conventional local hospital systems cannot match. See the GEMSS project web site at http://www.gemss.de/

Virolab is a Grid-based Virtual Laboratory for Decision Support in Viral Disease Treatment that benefits medical knowledge discovery and decision support for, e.g., HIV resistance. For more information see http://www.hoise.com/vmw/06/articles/vmw/LV-VM-02-06-39.html

The TATRC HealthGrid Integrated Research Team (IRT) is seeking to formulate a five-year research and development roadmap for the U.S. HealthGrid. The methodology for this Integrated Research Team is the pairing of biomedical and Grid expertise, to underscore the point that biomedical research can be accelerated and enhanced through collaborative partnerships and cooperation. See http://www.tatrc.org/website_healthgrid05/index.html

Finally, other healthcare grid activities of note mentioned in the recent HealthGrid 2006 proceedings included:

  • The ARTEMIS Project has developed a secure semantic Web service infrastructure for the interoperability of healthcare information systems.
  • The Health-e-Child Consortium is developing an integrated grid-based healthcare platform for European pediatrics.
  • The BIOPATTERN Grid test bed has been designed for bio-profiling to detect dementia and brain injury on an individual basis.

Benefits of Grids

There are many benefits to be accrued from the use of grids. These include:

Business benefits

  • Accelerate time to results
  • Enable collaboration and promote operational flexibility
  • Efficiently scale to meet variable business demands
  • Increased/improved productivity
  • Better leverage existing capital investments

Technology benefits

  • IT Infrastructure optimization
  • Increase access to data and collaborative solutions
  • Resilient, highly available IT infrastructure

While there are privacy and security issues that must be addressed when implementing grid solutions, many of the concerns raised in the earlier years have now been largely addressed.

Grids & Electronic Health Records (EHR) Systems

There are a number of ways that grids can potentially be used with electronic health record (EHR) systems over the coming years. For example, (1) computational grids can be used to solve large-scale research problems in healthcare, using the unused power of computer workstations of EHR systems in healthcare provider organizations; (2) data grids can be established that don't share computing power but instead provide a standardized way to securely exchange patient data internally and externally from EHR systems for data mining and decision support; and (3) collaborative grids can be built that let geographically dispersed users share medical information and work together on complex cases using patient data sets and clinical images maintained in EHR systems of multiple healthcare provider organizations.

Public Health, Disease Registries, and the NHIN Grid

Both federal and state health departments in the United States gather public health information on births and deaths, immunization and vaccines, environmental and occupational health, and have established a number of disease registries. Many states also have some form of automated biosurveillance systems in development, related to the national Public Health Information Network (PHIN) and the National Electronic Disease Surveillance System (NEDSS). Increased adoption of electronic health record (EHR) systems throughout the United States and around the world will further support the creation and use of computerized disease registries to capture and track chronic conditions. By 2020, public health information systems in the United States, such as disease registries, will be integrated into grids linked by the National Health Information Network (NHIN) that will utilize the Next Generation Internet (NGI) or Internet2.

Recommendations & Next Steps

  • Consider establishing a workgroup to identify functional requirements and/or potential uses of grid systems for use by your healthcare organization.
  • Conduct a detailed literature search annually and obtain lessons learned from health grid projects underway at other institutions.
  • Identify potential organizations to collaborate with on the research, development, testing and use of grids in healthcare, e.g. medical schools, vendors.
  • Conduct a feasibility study into the use of grids and select potential pilot projects.
  • Investigate changes in business and IT processes that may need to be made in anticipation of utilizing grid technology.
  • Initiate and fund pilot project(s) and complete a detailed cost benefit analysis.
  • At the national level, consider establishing a position of "Health Grid Coordinator".

The authors delve deeper into the subject of Grids and other innovations and emerging technologies in healthcare in their upcoming book entitled Medical Informatics 20/20 to be published by Jones & Bartlett in January 2007. See http://www.jbpub.com/catalog/0763739251

Authors

Douglas Goldstein is a "Practical Futurist", Author and President of Medical Alliances, Inc. He guides leading healthcare organizations in clinical and business performance improvement through intelligent use of technology, knowledge management and "Distinctive Innovation". He can be reached at doug@medicalalliances.com

Peter Groen was the former Director of the Health IT Sharing (HITS) program within the Veterans Health Administration of the U.S. Department of Veterans Affairs. He recently retired and is now on the faculty of the Computer & Information Sciences Department at Shepherd University in West Virginia. He can be reached at pgroen@shepherd.edu


Peter Groen, Douglas Goldstein

[Medical IT News][Calendar][Virtual Medical Worlds Community][News on Advanced IT]