The move toward personalized medicine, in concert with the recent advances in computing, data acquisition, processing and interpretation, is transforming diagnostic and interventional medicine from a traditional artisanal craft based on clinicians? experience into a discipline that relies on objective decision-making based on the integration of multi-dimension and multi-modal data from heterogeneous sources. Computer- integrated diagnostic and interventional data science encompasses the processing, analysis, and interpretation of images and signals to improve the quality of a diagnostic or therapeutic goal. Improvements result from helping clinicians better diagnose disease, predict clinical outcome, better plan, deliver and monitor therapy, as well as advance training and simulation. Despite advances in computer-integrated diagnosis the therapy during the past decade, there has been a delay in introducing large-scale data science techniques into diagnostic, and especially interventional medicine. Although these disciplines have been transformed by the emergence of digital imaging (i.e., histology, pathology, and microscopy), miniature cameras (i.e., endoscopy, and multi- modality medical imaging to ?see? inside the human body, the seamless, wide-spread integration of computer- aided tools as part of the routine diagnostic and surgical environment has been slow. This delay has been attributed to the limited availability of diagnostic and interventional data science techniques that can robustly handle the size, diversity and dimensionality of the acquired data that must be manipulated, often in real time. Ongoing projects in my lab have focused on the development and validation of image-based computing, modeling, and visualization frameworks that 1) help clinicians quantify and track imaging biomarkers to diagnose and monitor disease progression, 2) identify and plan optimal therapeutic routes, and 3) guide, monitor, and deliver therapy under less invasive conditions. These tools have been developed and demonstrated primarily in the context of cardiac applications, orthopedic, lung, brain, and spine applications, in close collaborations with clinicians and industry partners. The long-term vision of the proposed program is to further advance computer-integrated diagnostic and therapeutic data science by continuing the development and validation of new techniques for biomedical computing and visualization. We will leverage our successes and extend our existing computing infrastructure to operate on a wider range of digital data. Their output will supply clinicians with the necessary visualization for diagnostic and therapeutic decision making across different tissues and organs. We will make the developed techniques available to the biomedical research and community whose research necessitates using image-based modeling, simulation, and visualization, as well as to clinician scientists who can promote their clinical translation. This research program will yield innovative biomedical computing and visualization tools that rely on standard-of-care biomedical data and cater to a broad range of minimally invasive diagnosis and therapy applications.
My vision and goal for the proposed research program is to contribute to the development and validation of a robust computing infrastructure for biomedical research in support of computer-integrated diagnostic and therapeutic data science. These tools will operate on standard-of-care multi-modality and multi-dimensional data to answer fundamental questions underlying health and disease, while the proposed research will not only focus on novel methodologies, but also on their implementation using a more affordable, accessible manner. Their utilization has the potential to change the paradigm of traditional disease diagnosis and treatment, assisting clinicians with the early and accurate disease diagnosis, identification of the most appropriate course of treatment, and the monitoring and delivery of less invasive, more accurate and more precise therapy.
