The Cancer Institute of New Jersey (CINJ) and Rutgers Univ. have entered into a collaborative research effort this week to develop diagnostic tools which can improve the accuracy of predicting patients' responses to treatment and related clinical outcomes. Through the use of advanced computer and imaging technologies that facilitate comparisons of cancerous tissues, cell and radiology studies, researchers and physicians expect to determine more accurate cancer prognoses, more personalized therapy planning and, subsequently, the discovery and development of new cancer drugs.
This new project is an extension of the "Help Defeat Cancer" (HDC) project in which IBM's World Community Grid was used to demonstrate the effectiveness of characterizing different types and stages of disease based upon the underlying staining patterns exhibited by digitally imaged cancer tissues. World Community Grid is a virtual supercomputer that gains its resources by thousands of volunteers donating their unused computer time.
Leveraging the experimental results gathered during the course of the HDC project, the team has recently received a $2.5-million grant through competitive funding from the National Institutes of Health. The central objective of this project is to build a deployable, grid-enabled decision support system to help researchers, physicians and scientists to automatically analyze and classify imaged cancer specimens with improved accuracy. It is intended to be a useful tool for supporting the selection of personalized treatments for people with cancer based upon how patients with similar protein expression signatures and cancers have reacted to treatments.
The team is expanding the first phase of the project that studied breast, colon and head and neck cancers to include other cancers as well. From the World Community Grid project, CINJ created a reference library of expression signatures and demonstrated a reliable means for performing high-throughput analysis of tissue micro-arrays.
In addition, investigators at CINJ also are establishing a Center for High-Throughput Data Analysis for Cancer Research that will tap into state-of-the-art computing resources and a Shared University Research Award provided by IBM. The primary objective of the Center is to develop pattern recognition algorithms that can simultaneously take into consideration information contained in digitally archived cancer specimens, radiology images and proteomic and genomic data for improved assessment of disease onset and progression.
David J. Foran, director of the Center for Biomedical Imaging & Informatics at CINJ and professor of pathology and laboratory medicine at UMDNJ-Robert Wood Johnson Medical School, is the lead investigator for the project.
"World Community Grid enabled us to validate our imaging and pattern recognition algorithms and establish a reference library of expression signatures for more than 100,000 digitally imaged tissue samples. The overarching goal of the new NIH grant is to expand the library to include signatures for a wider range of disorders and make it, along with the decision support technology, available to the research and clinical communities as grid-enabled deployable software. Through the use of mirror sites at CINJ and Ohio State Univ., and with the support of the NCI-funded cancer Biomedical Informatics Grid (caBIG) program at NIH, we hope to deploy these technologies to other cancer research centers around the nation," says Foran.
Leiguang Gong of IBM's T.J. Watson Research Center is leading a team of experts in high performance medical imaging and informatics. In this venture, he and his colleagues at the IBM research and technical labs will collaborate closely with Foran's team at CINJ and investigators at Rutgers.
As part of the new center, IBM is donating P6 570 series class systems, which will provide additional computational power for the project. The center will utilize grid technology to provide access to the software and database to collaborating investigators at Arizona State Univ., the Ohio State Univ. and the Univ. of Pennsylvania School of Medicine. The consortium will serve as a network-based test bed for optimizing the software during iterative prototyping.
