Diverse Applications, Common Approaches? Multispectral and Hyperspectral imaging methods are widely used in diverse applications including airborne and satellite remote sensing, biological microscopy, and diagnostic biomedical imaging systems. There has been a corresponding growth in algorithms and architectures for processing and analyzing multi- and hyper-spectral images, and there is abundant opportunity for researchers to learn image processing problems, approaches, insights, and algorithm designs from other areas. The goal of this symposium is to provide an opportunity for researchers to share insights across disciplines and application areas. Impact: The impact of multi- and hyper-spectral imaging is already impressive. Satellite-based and airborne platforms are enabling remote sensing of changes, anomalies, and trends in agricultural, mineral, military, and civilian land-use patterns. Similarly, in the biomedical multi- and hyperspectral imaging methods are enabling the spatial distribution and dynamics of multiple molecular markers in a manner that preserves their relative spatial context, in living biological systems (in vivo or in vitro), and/or fixed tissue samples of biomedical interest (e.g., biopsies), and are finding applications in basic scientific discovery, drug screening, and biomolecular tissue profiling. Invitation: Researchers in signal processing and diverse application domains are invited to submit their contributions.
University of California, Los Angeles
Hyperspectral imagery is a challenging modality due to the dimension of the pixels which can range from hundreds to over a thousand frequencies depending on the sensor. Most methods in the literature reduce the dimension of the data using a method such as principal component analysis, however this procedure can lose information. More recently methods have been developed to address classification of large datasets in high dimensions. In this talk I review a family of efficient graph-based classification methods for hyperspectral imagery. Using the full dimensionality of the data, we consider a similarity graph based on pairwise comparisons of pixels. Rather than constructing the full graph, which is computationally prohibitive, I review two methods for approximating it - one that involves a low rank approximation of the graph Laplacian and another method that involves nearest neighbor sampling of pixels. The graph itself is segmented using a pseudospectral algorithm for graph clustering that requires information about the eigenfunctions of the graph Laplacian but does not require computation of the full graph. With at most a few hundred eigenfunctions, we can implement the clustering methods designed to solve variational problems for a graph-cut-based semi-supervised or unsupervised classification. We implement OpenMP directive-based parallelism in our algorithms and show performance improvement and strong, almost ideal, scaling behavior. The methods can handle very large datasets including a video sequence with over a million pixels, and the problem of segmenting a data set into a pre-determined number of classes or unknown number of classes.
Andrea Bertozzi is an applied mathematician with expertise in nonlinear partial differential equations and fluid dynamics. She also works in the areas of geometric methods for image processing, crime modeling and analysis, and swarming/cooperative dynamics. Bertozzi completed all her degrees in Mathematics at Princeton. She was an L. E. Dickson Instructor and NSF Postdoctoral Fellow at the University of Chicago from 1991-1995. She was the Maria Geoppert-Mayer Distinguished Scholar at Argonne National Laboratory from 1995-6. She was on the faculty at Duke University from 1995-2004 first as Associate Professor of Mathematics and then as Professor of Mathematics and Physics. She has served as the Director of the Center for Nonlinear and Complex Systems while at Duke. Bertozzi moved to UCLA in 2003 as a Professor of Mathematics. Since 2005 she has served as Director of Applied Mathematics, overseeing the graduate and undergraduate research training programs at UCLA. In 2012 she was appointed the Betsy Wood Knapp Chair for Innovation and Creativity. Bertozzi's honors include the Sloan Research Fellowship in 1995, the Presidential Early Career Award for Scientists and Engineers in 1996, and SIAM's Kovalevsky Prize in 2009. She was elected to the American Academy of Arts and Sciences in 2010 and to the Fellows of the Society of Industrial and Applied Mathematics (SIAM) in 2010. She became a Fellow of the American Mathematical Society in 2013 and a Fellow of the American Physical Society in 2016. She won a SIAM outstanding paper prize in 2014 with Arjuna Flenner, for her work on geometric graph-based algorithms for machine learning. Bertozzi is a Thomson-Reuters/Clarivate Analytics `highly cited' Researcher in mathematics for both 2015 and 2016, one of about 100 worldwide in her field. She was awarded a Simons Math + X Investigator Awardin 2017, joint with UCLA's California NanoSystems Institute (CNSI). Bertozzi was appointed Professor of Mechanical and Aerospace Engineering at UCLA in 2018, in addition to her primary position in the Mathematics Department. In May 2018 Bertozzi was elected to the US National Academy of Sciences. Bertozzi has served on the editorial boards of fourteen journals: SIAM Review, SIAM J. Math. Anal., SIAM's Multiscale Modeling and Simulation, Interfaces and Free Boundaries, Applied Mathematics Research Express (Oxford Press), Applied Mathematics Letters, Mathematical Models and Methods in the Applied Sciences (M3AS), Communications in Mathematical Sciences, Nonlinearity, and Advances in Differential Equations, Journal of Nonlinear Science, Journal of Statistical Physics, Nonlinear Analysis Real World Applications; and the J. of the American Mathematical Society. She served as Chair of the Science Board of the NSF Institute for Computational and Experimental Research in Mathematics at Brown University from 2010-2014 and previously on the board of the Banff International Research Station. She served on the Science Advisory Committee of the Mathematical Sciences Research Institute at Berkeley from 2012-2016. To date she has graduated 35 PhD students and has mentored over 40 postdoctoral scholars.
National Institutes of Health
Stroke is a leading cause of death in the United States and even when survivable it can lead to complex and widespread debilitating changes in the brain, with marked deterioration of quality of life and potentially long recovery times required to restore lost functions. While extensive stroke research conducted thus far has aimed to better characterize the complex processes mediating brain tissue damage and recovery after stroke, our full understanding of these processes and designing comprehensive drug treatments to elicit most beneficial clinical outcomes are still in their nascent stages. To further advance this field of research, we used a comprehensive systems cell biology biomarker screening approach by combining large-scale multiplexed fluorescence immunohistology in a rat brain model of focal ischemic injury with whole brain slide scanning using a customized multispectral imaging platform. We processed serial 10-micron thick whole brain coronal sections collected from male Sprague Dawley rats at 72 hours following focal brain ischemia induced by using a standard stroke model involving middle cerebral artery occlusion (MCAO). Sections were repeatedly probed and sequentially imaged using a panel of up to 50 fluorescent biomarkers relevant to cellular and molecular processes mediating neuroinflammation, neuroplasticity, neurogenesis, gliogenesis and angiogenesis in response to MCAO. Unique combinations of these biomarkers enabled a comprehensive identification and quantitation of all relevant cell types (neurons, astrocytes, oligodendrocytes, endothelial cells, microglia, immune cells, etc.) and their changing functional states (reactive, resting, apoptotic, proliferative, etc.) in the brain after injury. The results show dynamic and highly complex spatiotemporal changes in brain tissue remodeling and recovery after ischemic injury eliciting distinct cellular/molecular and specific niche responses that develop both proximally and distally to the site of injury. This work demonstrates the crucial need and a workable solution to apply comprehensive multiplex fluorescence biomarker screening and multispectral imaging to resolve complex systems biology of brain in response to stroke associated injury. The practical computational solutions pertaining to processing very large image datasets and multi-parametric computational image analysis of these datasets are currently in development. This work is supported by Intramural Research Program, National Institutes of Neurological Disorders and Stroke, National Institutes of Health.
Dr. Dragan Maric received B.S. degrees in biochemistry and microbiology in 1981 from University of Maryland, USA, and a M.S. in immunology in 1985 and a Ph.D. in neuroimmunology in 1989 from University of Belgrade, Yugoslavia. Dr. Maric completed his postdoctoral training in developmental neurobiology at the National Institutes of Neurological Disorders and Stroke (NINDS), National Institutes of Health (NIH) in Bethesda, Maryland, USA from 1991-1993, and in mucosal immunology at McMaster University, Hamilton, Canada from 1993-1994. After returning to NINDS in 1994, Dr. Maric worked in the Laboratory of Neurophysiology first as a visiting scientist from 1994-2000 and then as a staff scientist from 2000-2010 focusing his research primarily in the field of developmental neurobiology. He became Manager of the NINDS Flow and Imaging Cytometry Core Facility in 2001 with the primary mission to support basic, translational and clinical intramural research at NINDS and other institutes across NIH. Dr. Maric’s continued interests include researching the seminal properties of neural stem cells and their differentiating progeny during embryonic, postnatal and adult CNS development and adapting new methodologies to more effectively study the complex processes in brain systems biology under physiological and pathophysiological conditions.
|Wednesday, November 28|
|09:40 - 10:40|
|DL DL-MHI.1: Andrea Bertozzi: "Hyperspectral Image Classification Using Graph Clustering Methods"|
|11:00 - 12:00|
|DL DL-MHI.2: Dragan Maric: "Deciphering the complexity of brain system biology via whole brain multispectral imaging"|
|14:00 - 15:30|
|MHI-L.1: Multispectral and Hyperspectral Imaging and Analysis I|
|15:50 - 17:20|
|MHI-L.2: Multispectral and Hyperspectral Imaging and Analysis II|
Submissions are welcome on topics including:
Prospective authors are invited to submit full-length papers (up to 4 pages for technical content including figures and possible references, and with one additional optional 5th page containing only references) and extended abstracts (up to 2 pages, for paper-less industry presentations and Ongoing Work presentations).. Manuscripts should be original (not submitted/published anywhere else) and written in accordance with the standard IEEE double-column paper template. Accepted full-length papers will be indexed on IEEE Xplore. Accepted abstracts will not be indexed in IEEE Xplore, however the abstracts and/or the presentations will be included in the IEEE SPS SigPort. Accepted papers and abstracts will be scheduled in lecture and poster sessions.
|Paper Submission Deadline|
|Review Results Announced||September 7, 2018|
|Camera-Ready Papers Due||September 24, 2018|
|November 5, 2018||Hotel Room Reservation Deadline|