Sign In

Department of Biomedical Informatics   

Joel Saltz, MD, PhD, Chair
Created in 2001 under the leadership of Joel Saltz, MD, PhD, the Department of Biomedical Informatics (BMI) has grown to 13 faculty, 15 researchers and 14 students. Saltz and colleagues have brought international recognition to Ohio State in BMI, which applies computational and data-management techniques in biomedical data retrieval and integration, imaging, simulation, clinical/translational informatics, computational biology and comparative
genomics. These techniques are integrated in and supported by software tools and systems developed through advanced computer science methodologies. A specific research focus is
middleware systems to enable Grid computing. These efforts have led to innovations in: runtime support for distributed data processing and management; strategies for large-scale biomedical image analysis and bioinformatics computation; strategies for efficient querying and processing of data in distributed environments; grid-based security frameworks; and
tools to hide the complexity of developing grid applications. BMI has received much external funding support. It is also the lead site for the National Cancer Institute’s caGrid project.

Ongoing Research Programs

  • Bioinformatics and Systems Biology – BMI’s Bioinformatics and Systems Biology Program is focused on three areas: applying approaches using available data and deriving novel concepts that can be tested in the lab regarding comparative genomics of viruses, geographic and evolutionary mapping of infectious agents, and large-scale phylogenetic computation; applying systems biology approaches, such as co-expression analyses and proteomics combined with experimental validation in the lab, to identify biochemical steps in the process of a breast cell transforming into a cancer cell; and developing a comparative gene expression analysis method that allows researchers to identify commonly regulated genes in multiple types of cancer.
  • Clinical/Translational Informatics – BMI’s Clinical/Translational Informatics Program focuses on three areas: observation and modeling of workflow in clinical and research environments to determine strategies for optimizing and introducing information technologies; development of advanced methods and technologies to support multidimensional biomedical data acquisition, integration and presentation; and development of integrated information-technology platforms targeting clinical and translational research domains.
  • Imaging Informatics: BMI’s Imaging Informatics Program develops algorithms, tools and technologies to analyze large biomedical-image data, including algorithm development for
    using imagery as a phenotyping tool in systems biology, and algorithms for automated classification of neuroblastoma and follicular lymphoma. In imaging informatics, BMI is pursuing both radiological imaging analysis and computational histopathology efforts. In radiological image analysis, the Department is developing image-analysis techniques to
    predict radiation-treatment outcome for cervical cancer; in computational histopathology, it is working on neuroblastoma, follicular lymphoma, and the tumor microenvironment of
    breast cancer.
  • Multiscale and Grid Computing: This program develops middleware technology and techniques to enable management, sharing and manipulation of data at multiple scales
    across heterogeneous, dynamic collections of storage and computation systems. Some applications include: largescale, collaborative, biomedical clinical studies; analysis of
    gene expression and functional imaging information; imaging, analysis and simulation of subsurface physics and data-driven management of oil reservoirs; analysis of satellite data; and analysis of multiresolution, multiple-grid simulation data sets. The program targets techniques to support optimized distributed data storage, indexing, retrieval and processing of large datasets across many distributed storage systems; these techniques are integrated in systems software able to leverage knowledge of descriptive metadata in a way that supports many applications.

Research Accomplishments of 2007

  • The Laboratory for Translational Research Computing (LTRC), led by Philip Payne, PhD, established itself as a major contributor
    to the translational research efforts of numerous Medical Center investigators and their external collaborators. LTRC investigators’ efforts yielded such software as a hypothesisgeneration engine that can be applied to existing phenotypic and biomolecular data sets, advanced clinical trials management tools, ontology-anchored abstraction layers for complex data repositories such as data warehouses, and natural language processing platforms to extract structured data from clinical narrative text, such as that found in discharge summaries and pathology reports. These efforts were collaborations with several departments and research consortia, including Ohio State’s James Cancer Hospital and Solove Research Institute, the College of Medicine’s Department of Internal Medicine, the Chronic Lymphocytic Leukemia Research Consortium, and the Osetoarthritis Initiative.
  • LTRC members were recognized for outstanding research, expertise and contributions to the Biomedical Informatics community. Philip Payne, PhD, was selected to co-chair the
    American Medical Informatics Association (AMIA) Clinical Research Informatics Steering Task Force, and several other members were invited to present their work at national meetings such as those sponsored by AMIA and the National Science Foundation. In addition, LTRC members played a pivotal role in development of Ohio State’s groundbreaking Center for Clinical and Translational Science, including pursuit of the prestigious Clinical and Translational Science Award from the National Institutes of Health.
  • A team led by Dan Janies, PhD, created an interactive genomic and geographic map using phylogenetic software and GoogleTM Earth to reconstruct evolution and spread of H5N1 influenza lineages over the past decade. By examining an H5N1 phylogeny projected onto the globe and using character-evolution techniques, the team also studied, both visually and statistically, whether key genotypes in various viral proteins are correlated with host shifts.
  • BMI announced the creation of the Software Research Institute (SRI), which enables the design, development and release of high-quality software and middleware. Software produced through the SRI maintains high priority in design, architecture, quality, documentation and training; it also exceeds community standards for usability and quality in open source projects while remaining focused on scientifically motivated research projects relevant to BMI. SRI is led by co-directors Shannon Hastings, Stephen Langella and Scott Oster, and it manages the caGrid and Cardio Vascular Research Grid (CVRG) projects.
  • The latest version of caGrid, version 1.1, was released in September 2007. The cancer-research community recognized the latest version for its improvements in user-requested areas such as usability and security. NCI’s caBIG Web site has credited Ohio State’s BMI Department for its work as the lead supporting institution for the development of caGrid 1.1; the BMI caGrid team received caBIGTM awards two years in a row for leadership and contributions to developing caGrid infrastructure.
  • In imaging, the success of BMI students has been recognized at prominent conferences. A paper titled “Outcome Prediction for Radiation Treatment of Cervical Cancer by Assessing Tumor Heterogeneity and Temporal Change” by MD/PhD student Jeffrey Prescott was selected as one of the best submissions to the Society for Imaging Informatics in Medicine Conference, and he was awarded a travel grant. Another paper, “Neuroblastoma Stroma Classification on the Sony Playstation 3” by PhD student Timothy Hartley won the Best Scientific Session in Telemedicine, High-Performance Computing and Imaging Informatics Award, while BMI’s In Vivo Imaging Core Middleware Development Team received an “Embodying the Vision Award” from the NCI’s caBIG™ program.
  • Umit Catalyurek, PhD, earned from the National Science Foundation a five-year Faculty Early Career Development Award for “Scalable Combinatorial Scientific Computing,” a project in which he will design tools to accelerate the analysis of massive datasets that can be untangled and understood only by using powerful computers and supercomputers. The laboratory of Jeffrey Parvin, MD, PhD, developed biochemical assays for the function of the breast cancer tumor-suppressor gene called BRCA1, found that its enzymatic activity regulates synthesis of mRNA, discovered how this enzyme controls the function of an organelle known as the centrosome, and developed a systems approach of analyzing co-expressed genes to find another protein that provides critical regulation of the centrosome.