NSF-Funded Researcher Seeks to Acquire Medical Images Faster and Diagnosis Disease Earlier
The National Science Foundation has awarded Mathews Jacob, Ph.D., assistant professor in the Department of Biomedical Engineering at the University of Rochester, its CAREER Award. The five-year award will fund his research into the design of computer programs that accelerate the capture of high-resolution images, and that promise to enable early diagnosis of cancer in particular. The team’s goal is to help realize the potential of imaging technologies like magnetic resonance imaging (MRI), and to make sought-after clinical applications possible for the first time.
The National Science Foundation (NSF) is an independent federal agency that funds about 20 percent of basic research conducted by America's universities, with one focus being mathematics and computer science. The Faculty Early Career Development (CAREER) Program awards grants to junior faculty who exemplify the dual teacher-scholar role, and those promise to become lifetime leaders. The $399,600 grant began on July 1, 2009.
Jacob’s award-funded project is titled "Efficient Image Sparsifying Operators: Theory, Algorithms and Applications." The goal of the project is to develop algorithms that enable the compression of high-resolution images. Compressed data can be acquired faster. Certain valuable techniques are currently unrealistic because they generate too much data for current systems to handle. To get clear images, they would require patients to remain perfectly still inside scanners for long periods of time. Jacob’s work has the potential to considerably shorten such scans.
He is especially interested in designing new methods to speed up MRI, an imaging technology that uses a magnet and radio frequency waves to create images of organs and soft tissues with no exposure to radiation. A related technique, MR spectroscopic imaging captures chemical information about the organ, while routine MRI only captures its shape. The potential of MR spectroscopy lies in the fact that cancer can be detected by imaging related chemical imbalances long before a tumor grows big enough to change the shape of organs. Unfortunately, gathering information on trace amounts of disease-related chemicals with current technologies requires prohibitively long scans.
To realize the promise of these technologies, researchers need to improve the current mathematical techniques used to reduce the redundancy in natural images. Images are captured in pixels, the dots that make up a picture. At the heart of Jacob’s computational methods are new techniques for reducing the amount of information needed to accurately re-create an image.
While older techniques must capture data on every pixel in a picture, Jacob is designing programs (transforms, operators) that recognize many pixels may have the same value. By reducing redundancy, one can drastically reduce the amount of information to be collected, which profoundly accelerates the speed of acquiring images.
In related work, Jacob leads a project titled "Model-based MR Spectroscopic Imaging for Brain Cancer Treatment Planning", which is funded by the Clinical and Translational Science Institute at the University of Rochester Medical Center. His laboratory, the Computational Biomedical Imaging Group, is located within the Rochester Center for Brain Imagining. One purpose of these centers is to gather functional MRI (fMRI) data, which shows where blood rushes to in the brain as a person performs a given task.
“Currently functional MRI scans are restricted to a low resolution images,” Jacob said. “We are working on technologies that will enable researchers to clearly see which brain structures turn on as our eyes capture a picture, as we move our limbs, as we have a thought or feel an emotion, in terms of a two-second blood rush. Related insights will improve our basic understanding of brain function, launch new treatment approaches for major diseases and empower efforts to restore sight and movement to the disabled.”
“It is great to see the quality of Mathews' work recognized in this way,” said Richard Waugh, Ph.D., chair of the Department of Biomedical Engineering at the University of Rochester. “These awards are the gold standard for identifying top young faculty in engineering fields and it is a great honor to be selected.”