David Fuentes
Assistant Professor
The University of Texas MD Anderson Cancer Center
Department of Imaging Physics
My primary research interests concern the development, implementation, and validation of high performance human-assisted computational tools for MR-guided thermal therapies. The underlying research philosophy being that computational science, along with traditional theory and experiment, is a core and essential third pillar of the scientific discovery and engineering design process. Given the trends of increasingly powerful computational and visualization resources and exascale computing abilities on the horizon, computational science will continue to have an escalating role in providing more optimal planning, targeting, monitoring, and assessment of image-guided procedures. The unique dynamic closed loop control system, facilitated by the coupling of the predictive capabilities of computational simulation with real-time imaging feedback, has the potential to enable novel and robust model-constrained approaches to imaging as well as to lay the foundation for reliable minimally invasive computer-assisted treatment modalities. Such technology could dramatically increase treatment safety and efficacy as well as reduce associated treatment morbidities. Within this paradigm, the expertise of computational scientists is becoming indispensably important to anticipate and understand the underlying architectures of forthcoming computing hardware, develop the software to realize the potential of these machines, and obtain meaningful results that advance the field.
Project/Techniques:
Tutorial material will expose students to advanced mathematical modeling of various aspects of image-guided interventions ranging from the imaging acquisition to therapy delivery. Students will formalize mathematical models from first principles and develop appropriate numerical solutions. Numerical methods will include aspects of inverse problem solutions, finite element modeling, processing of MR temperature imaging data, quantifying uncertainty in computer predictions, and optimization. The methodology and experience will be applicable to a range of application areas outside the particular tutorial material.
Education & Training
Ph.D. - The University of Texas at Austin - 2008