Prof Xu is one of the world’s leading authorities on computational radiation dosimetry involving Monte Carlo simulations and human phantoms — models of the human body for the purposes of quantifying radiation dose to organs. His ground-breaking research activities helped advance the field of radiation dosimetry. His research approaches are interdisciplinary and transdisciplinary, involving information technology, nanomaterials, terahertz, 3D printing, cancer stem cells, compressive sensing, and, recently, high-performance computing devices such as GPUs and MIC coprocessors.
The methods and tools developed by Prof Xu have been widely used to solve problems in many different fields. His most important research achievements can be summarized in the following 3 areas:
(1) Protection of Nuclear Workers
Nuclear power is a clean, efficient, and lasting solution to the increasing demand for energy. Radiation sources are also widely at industrial and scientific facilities. The safety of workers who handle various radioactive materials and radiation-emitting devices is therefore a top priority in the workplace. Prof Xu is credited for developing the so-called “Two Dosimeter” methodology for measuring personnel doses and the method has been adopted by the nuclear power industry in the U.S. He developed two software packages, EDECalculator and VRDose Simulator, which were distributed by EPRI to the nuclear power industry to monitor radiation to the workers. Prof Xu pioneered advanced human models that significantly improved the accuracy of radiation protection dosimetry. In the past 20 years, his group reported techniques for advanced human phantoms (such as VIP-Man, RPI-Pregnant Female, RPI-Adult Male and Female, and RPI-Obese Patients). The paper by Prof Xu and his two PhD students on the VIP-Man is a widely cited paper in the field and the model has been freely distributed to hundreds of users worldwide (RSICC.ornl.gov). Prof Xu is a founder and organizer of several scientific groups and conferences. His review articles and book chapters are frequently cited.
(2) Medical Diagnostic Imaging and Cancer Radiotherapy
Computed Tomography (CT) — a technology uses rotating X-ray to project images — is the most widely used 3D/4D tomographic imaging modality today, but it is now known to pose a potential risk without careful management. Prof Xu developed cutting-edge technologies to accurately track radiation to patients of various ages and sizes. Prof Xu founded a company to translate his research at Rensselaer into a commercial product, a software package called VirtualDose. Developed with grant support from the NIH, VirtualDose now has more than 40 licensed radiology users worldwide. Prof Xu is expanding the research into other imaging modalities such as interventional radiology and PET/CT.
High energy X-rays, electrons and protons are also used in different forms of cancer treatment. Prof. Xu developed one of the first “respiration-simulating” methods to minimize the chance of missing the targets during external-beam treatment. Prof Xu was also one of the first to perform dosimetric studies about “secondary cancer after radiotherapy” — a cause of failure in radiation treatment — and, through a series of conferences and papers including a consensus report that Prof Xu co-authored with 11 leading radiation oncologists, Prof Xu has established himself as one of the international authorities on that topic.
Taking advantage of a collaborative environment at Rensselaer, Prof Xu has published papers or written grants on topics such as: 3D image registration and visualization (with C Stewart in comp sicence and R Radke in electric eng), terahertz imaging (with XC Zhang in physics - now at Rochester), nanotube-based brachytherapy (T Lu in physics and P Ajan in materials science - now at Rice), respiration-motion modeling (with S De in mech eng), compressive sensing (with B Yazici in electric eng) and parallel computing (with C Carothers in comp science, M Shephard in SCOREC, and W Ji in nucl eng).
In the past 20 years, Prof Xu collaborated externally with as many as 10 medical centers. Serving as a PI on these projects, Prof Xu’s research in this category have been continuously funded by NIH, resulting in innovative and transdisciplinary solutions to urgent healthcare problems.
(3) High Performance Computing
Exa-scale parallel computing capabilities are expected to arrive near the end of this decade. The Graphics Processing Unit (GPU) and Many-Integrated-Core (MIC) technologies are already revolutionizing scientific computing and gaming, with unprecedented parallelization and data communication algorithms. However, “heterogeneous parallel computing” is a relatively young field of research and the scientific computing community at large has not been fully prepared for the emerging technologies. Many “legacy codes” in nuclear engineering originated from the Manhattan Project in 1940s now run the risk of being handicapped when hardware architectures are incompatible with the traditional CPU-based programming paradigm. In early 2015, the Department of Energy announced specific plans to construct new supercomputers involving the latest GPU and MIC technologies.
In anticipating the trends in parallel computing, Prof Xu forged a consortium, in 2010, with colleagues at RPI (C Carothers in computer science, M Shephard in SCOREC, and W Ji in nuclear eng). They set out to tackle a long-known challenge in nuclear engineering — the Monte Carlo radiation transport simulation, although the most accurate and the only way to handle heterogeneous and 3D media, is painfully slow. The team envisioned a revolutionary new Monte Carlo computing capability of near “realtime” speed that is only possible with future technologies.
Since 2010, Prof Xu led this team which also included one postdoc and three PhD students to successfully attract one major grant from the NIH and to publish 3 journal articles and 20 conference papers. Thus far, the team has accomplished the following: (1) Developed the framework of ARCHER — a new Monte Carlo radiation transport software designed specifically to run on different generations of CPU, GPU, and MIC, each device requiring a different set of software design. (2) Demonstrated that a traditional CT imaging dose calculation that took 4 hours can now be computed in 7 seconds, (3) Demonstrated that a traditional cancer treatment plan that took 24 hours can now be computed in 60 seconds. (4) Demonstrated for the first-time the feasibility of concurrent execution of a Monte Carlo algorithm on CPU, GPU, and MIC. (5) Graduated two PhD students and trained several new students. (6) Developed strategic plan to take advantage of both inter-node and intra-node programming models (MPI, openMP, pThreads, CUDA) for the next-generation DOE supercomputers. (7) Submitted a number of major grant applications to DOE and NIH. The group is rapidly becoming a leader in that new research field and Prof Xu has been invited to chair the session: “High-Performance Computing and Algorithms for Advanced Architectures” at Supercomputing in Nuclear Applications and the Monte Carlo Method Conference in 2015. Since Monte Carlo computing is at the core of computational nuclear engineering and physics, Prof Xu and his team is now positioned at the forefront of an exciting research field in the next 10 years.
HONORS and AWARDS
For his research activities, Prof Xu has received numerous awards including: NSF/CAREER Award, Rensselaer School of Engineering Excellence in Research, Rensselaer School of Engineering Outstanding Team Award, American Nuclear Society Best-Paper Award, Journal of Physics in Medicine and Biology Best Paper Award, American Nuclear Society Radiation Protection and Shielding Division Professional Excellence Award, and the Randal S. Caswell Award for Distinguished Achievements.
In the past 20 years, Prof Xu has received a total of about $15M research funding from NSF, DOE, NASAS, NIH, NIST, EPRI and nuclear industry (Y13 annual expenditure was $1.04M).
Prof Xu has authored 7 book/monograms, 120 peer-reviewed journal papers/chapters, 300 conference abstracts, 100 invited seminars/plenary presentations, 5 patents/disclosures, and 6 software packages. His papers are published in top journals and rank among the most highly cited articles in the radiation safety and medical physics research literature.
In the past 20 years, Prof. Xu has also received more than 100 pieces of media coverage by newspapers and TV stations, including the Wall Street Journal.
TEACHING AND MENTORING
In the past 20 years, Prof Xu has taught courses from the freshman to graduate levels and was regarded by some students as their favorite instructor. He has supervised 17 undergraduate research/design projects – out of which 5 received Best Project Awards. Prof Xu has graduated 18 Ph.D. and 11 M.S. students — among them, 10 have received professional society graduate fellowships and 7 are now faculty members (including the head, Department of Physics, US Military Academy).
Over the years, Prof Xu has developed an effective teaching and mentoring system. Many undergraduate students made career decisions after having taken Prof Xu’s classes. Graduate students who are now faculty members often use Prof Xu as a model in mentoring.
SERVICE TO RENSSELAER
From 1994 – 2000, Prof Xu served as the Institute Radiation Safety Officer, responsible for nuclear and radiation safety on the campus. In the past 20 years, he has held increasing leadership responsibilities in the nuclear engineering program within the MANE Dept. He served over 10 years as the program’s ABET Director and successfully completed the exhaustive degree accreditation processes in 2007 and 2013. From 2011 to 2013, he served a full two-year term as the head of the nuclear eng program which consists of 10 full-time faculty members and 160 students who are supported by two large facilities — the Gaerttner Linear Accelerator (LINAC) and the Walthousen Reactor Critical Facility (RCF). The program’s graduate enrollment increased by 28% and the research expenditure increased by 43% compared to 2011. Under his leadership, the program significantly elevated its national visibility through the following initiatives:
• strategic planning,
• new website,
• visitation by 10 department heads,
• representation of RPI at national events,
• nomination of colleagues for internal/external awards, and
• establishment of two endowed funds (The Block Prize for Best PhD Research and the Lahey Distinguished Lectures).
In these activities, Prof Xu gained respect from his colleagues for his leadership skills and his dedication to the nuclear engineering education.
NATIONAL AND INTERNATIONAL LEADERSHIP
As one of the pioneers in the field of computational radiation dosimetry using human phantoms, Prof Xu has written widely cited review articles and book chapters about future research directions. He serves on 5 journal editorial boards. He was president of Council on Ionizing Radiation Measurements and Standards (CIRMS) that sets national priorities on radiation standards involving leaders from government, academia and private-sector industry. For his contributions, Prof Xu was selected to receive the Randal S. Caswell Award for Distinguished Achievements by CIRMS. He served as the chair of ANS/Radiation Protection and Shielding Division (1200 members). He has given 100 invited seminars / plenary /keynote talks. He co-founded the International Consortium of Computational Human Phantoms. In 2008, He was elected to a 6-year term (and was re-elected in 2014) as a member of the National Council of Radiation Protection and Measurements (NCRP) that is made of 50 nation’s top experts in this field. He co-edited a book “Handbook of Anatomical Models for Radiation Dosimetry” with 60 authors from 13 countries. He organized the 3rd and 4th International Workshop on Computational Phantoms for Radiation Protection, Imaging and Radiotherapy. In 2012, Prof Xu was appointed a three-year term on US Environmental Protection Agency’s Science Advisory Board Radiation Advisory Committee. Prof Xu is an elected fellow of American Nuclear Society, American Association of Physicists in Medicine and Health Physics Society.