The design of successful numerical methods for image processing and other biomedical applications require a thorough understanding of the underlying problems and of the desired solutions, as well as of the instrumentation for data collection and the computational environment. This proposal seeks support for the PI to spend his sabbatical year, and part of the summers attached to it at the Biomedical Engineering department at Case Western Reserve University. J. Duerk and D. Wilson will serve as hosts to the PI during his stay. A year of full time immersion in the biomedical environment at CWRU will give the PI an opportunity to strengthen and broaden his image processing knowledge and to view the field from the engineering perspective. He will also acquire significant insight in various biomedical engineering applications whose solution is computationally challenging. The PI's expertise on numerical methods for large-scale computations and the recent applications to the restoration of images will make it possible to begin interacting with faculty and students in the hosting department from the start, thus allowing him to take advantage of all learning opportunities. It is envisioned that this interaction will develop into a solid and fruitful cooperation on the development of better numerical methods for novel medical applications, e.g., interventional MRI for a new less invasive cancer treatment. The new knowledge and skills will significantly broaden the PI's expertise and will open new avenues for his future research. Upon returning to his home institution, where there is neither an engineering school nor a medical school, he will divulge his new knowledge of medical imaging and of the engineering perspective on it by directing the work of Masters and PhD students, designing new, interdisciplinary courses and giving a series of lectures and seminars. Upon return to his department the PI will be able to -introduce applications from biomedical enginee ring in graduate and undergraduate courses on numerical analysis and scientific computing -assign computational problems arising in biomedical engineering for Ph.D. and Masters theses, possibly in collaboration with members of the Department for Biomedical Engineering at CWRU. This should yield thesis topics that students would find interesting and attractive, as well as make the students more attractive on the job market -organize seminars on computational problems in biomedical engineering, and if there is interest, develop courses on "Computational Problems in Biomedical Engineering" -serve as liaison between his department and the Department of Biomedical Engineering at CWRU. This IGMS project is jointly supported by the MPS Office of Multidisciplinary Activities (OMA) and the Division of Mathematical Sciences (DMS).
