Dr. Schubert is a full professor in the Electrical and Computer Engineering Department at Baylor University. His interest areas are the fast, numerically stable solution of large, sparse problems, particularly for medical imaging and treatment; the design of electrical and embedded systems for medical and biological applications; the patterned growth of extremeophiles and the search for life in space; and the mathematical modeling of synchronization. Dr. Schubert has been a team leader on the December 2012 National Geographic expedition to Cueva de Villa Luz in Mexico and his work made the cover article of National Geographic July 2013. Dr. Schubert has worked on medical projects with Loma Linda University (LLU), LLU Medical Center, University of California San Francisco, Baylor Scott and White Medical centers in Dallas and Temple, and holds an adjunct full professor position at LLU. Dr. Schubert's medical projects range from developing proton computed tomography, to functional near infrared spectroscopy (fNIRS) brain monitoring to artificial intelligence for colonoscopy. He has more than 80 peer reviewed publications.
Industry Expertise (2)
Areas of Expertise (13)
Imaging and 3d Reconstruction
Numerical Linear Algebra
Robust and Uncertain Systems
Bio-medical Applications of Computer Science and Engineering
University of California, Santa Barbara: Ph.D., Electrical and Computer Engineering, Control Systems 2003
Minors: Applied Mathematics and Signals and Communication
Dissertation: A New Look at Estimation
University of California, Los Angeles: M.S., Electrical Engineering, Control Systems 1992
Minors: Applied Mathematics and Digital Signal Processing
Project: Guidance and Control of Amateur Rockets
University of Redlands: B.S., General Engineering 1991
Project: Solar Desalination
Honors: Dean’s List (multiple times), Cum Laude
- IEEE - Senior Member
- IEEE Computer Society - Member
- SIAM - Member
- Bioinformatics Organization, Inc. - Member
- International Association of Engineers - Member
- Golden Key Honor Society - Member
- Gamma Lambda Chapter of Phi Beta Delta, the Honor Society for International Scholars - Member
Media Appearances (3)
Be safe when capturing the eclipse
Fox 44 News online
We will all want to take pictures and video of the solar eclipse. But do you know how to get those pictures without damaging your camera or phone?
One of the most important things to know is to not use a DSLR camera, where you have to look through an eye piece. Those cameras reflect light back into your camera, where it can cause eye damage.
Professor provides tips to photograph eclipse
KCEN-TV Ch. 6 (Waco, Temple, Killeen/NBC) tv
Keith Schubert, Ph.D., professor of electrical and computer engineering in Baylor’s School of Engineering and Computer Science, provides tips to viewers who hope to take photos of the Aug. 21 solar eclipse. Schubert warns that failure to take proper precautions could damage viewers’ eyes or their cameras.
Baylor Professor featured in National Geographic for Work in Extreme Environments
Baylor Media Communications online
When looking at the series of photos on Keith Schubert’s computer screen, most people will likely see what they believe is some sort of black goo arranged like an intricate maze on a rock wall.
But Schubert, Ph.D., associate professor of electrical and computer engineering in Baylor University’s School of Engineering and Computer Science, sees life.
Specifically, he sees patterns of life.
Ernesto Gomez, Zongqi “Ritchie” Cai, Keith Evan Schubert
We propose a definition of parallel state, derive a phase space from this state, and calculate the entropy of states and full executions using combinatorial analysis. A main contribution of this work is the introduction of an experimentally measurable phase space, which we then use to analyze execution states, ensembles of states, and ensembles of complete executions. We show that the entropy analysis reveals both expected and unexpected features of execution, and application of principal component analysis shows capability to extract execution details at the level of individual process states, as well as reveal hardware properties such as network or memory communications.
K.E. Schubert, et al.
We report on the operation and performance tests of a preclinical head scanner developed for proton computed tomography (pCT). After extensive preclinical testing, pCT is intended to be employed in support of proton therapy treatment planning and pre-treatment verification in patients undergoing particle-beam therapy. In order to assess the performance of the scanner, we have performed CT scans with 200 MeV protons from both the synchrotron of the Loma Linda University Medical Center (LLUMC) and the cyclotron of the Northwestern Medicine Chicago Proton Center (NMCPC). The very high sustained rate of data acquisition, exceeding one million protons per second, allowed a full 360° scan to be completed in less than 7 min. The reconstruction of various phantoms verified accurate reconstruction of the proton relative stopping power (RSP) and the spatial resolution in a variety of materials. The dose for an image with better than 1% uncertainty in the RSP is found to be close to 1 mGy.
Keith E, Schubert, et al.
Image registration techniques based on anatomical features can serve to automate patient alignment for intracranial radiosurgery procedures in an effort to improve the accuracy and efficiency of the alignment process as well as potentially eliminate the need for implanted fiducial markers. To explore this option, four two-dimensional (2D) image registration algorithms were analyzed: the phase correlation technique, mutual information (MI) maximization, enhanced correlation coefficient (ECC) maximization, and the iterative closest point (ICP) algorithm. Digitally reconstructed radiographs from the treatment planning computed tomography scan of a human skull were used as the reference images, while orthogonal digital x-ray images taken in the treatment room were used as the captured images to be aligned. The accuracy of aligning the skull with each algorithm was compared to the alignment of the currently practiced procedure, which is based on a manual process of selecting common landmarks, including implanted fiducials and anatomical skull features. Of the four algorithms, three (phase correlation, MI maximization, and ECC maximization) demonstrated clinically adequate (ie, comparable to the standard alignment technique) translational accuracy and improvements in speed compared to the interactive, user-guided technique; however, the ICP algorithm failed to give clinically acceptable results. The results of this work suggest that a combination of different algorithms may provide the best registration results. This research serves as the initial groundwork for the translation of automated, anatomy-based 2D algorithms into a real-world system for 2D-to-2D image registration and alignment for intracranial radiosurgery. This may obviate the need for invasive implantation of fiducial markers into the skull and may improve treatment room efficiency and accuracy.
Ernesto E. Gomez, Keith Schubert
Modern multiprocessor architectures have exacerbated problems of coordinating access to shared data, in particular as regards to the possibility of deadlock. For example semaphores, one of the most basic synchronization primitives, present difficulties. Djikstra defined semaphores to solve the problem of mutual exclusion. Practical implementation of the concept has, however, produced semaphores that are prone to deadlock, even while the original definition is theoretically free of it. This is not simply due to bad programming, but we have lacked a theory that allows us to understand the problem. We introduce a formal definition and new general theory of synchronization. We illustrate its applicability by deriving basic deadlock properties, to show where the problem lies with semaphores and also to guide us in finding some simple modifications to semaphores that greatly ameliorate the problem. We suggest some future directions for deadlock resolution that also avoid resource starvation.