config.options.chkHttpReadOnly = true;
// Eric Shulman - ELS Design Studios\n// "Mixed HTML and wiki-style rendering" Plug-in for TiddlyWiki version 1.2.25 or above\nversion.extensions.HTMLFormatting = {major: 1, minor: 0, revision: 1, date: new Date(2005,7,26)};\nwindow.coreWikify=window.wikify;\nwindow.wikify = function(tiddlerText,theViewer,highlightText,highlightCaseSensitive)\n{\n var startHTML = tiddlerText.indexOf('<'+'html'+'>');\n var endHTML = tiddlerText.lastIndexOf('<'+'/html'+'>');\n if (startHTML==-1) // bypass HTML parsing\n { coreWikify(tiddlerText,theViewer,highlightText,highlightCaseSensitive); return; }\n if (startHTML>0) // wikify everything up to HTML tag\n coreWikify(tiddlerText.substr(0,startHTML-1),theViewer,highlightText,highlightCaseSensitive);\n if (startHTML!=-1) // browser parse everything between HTML and /HTML tags (or end of text)\n {\n var HTMLText = tiddlerText.substr(startHTML);\n if (endHTML!=-1) HTMLText = tiddlerText.substring(startHTML,endHTML+7);\n // suppress wiki-style literal handling of newlines\n if (HTMLText.indexOf('<hide linebreaks>')!=-1) HTMLText=HTMLText.replace(regexpNewLine,' ');\n // strip any carriage returns added by Internet Explorer's textarea edit field\n HTMLText=HTMLText.replace(regexpCarriageReturn,'');\n // encode newlines as \sn so Internet Explorer's HTML parser won't eat them\n HTMLText=HTMLText.replace(regexpNewLine,'\s\sn');\n // encode macro brackets (<< and >>) so HTML parser won't eat them\n HTMLText=HTMLText.replace(/<</g,'%macro(').replace(/>>/g,')%');\n // create a span to hold browser-parsed DOM objects\n var newSpan = createTiddlyElement(theViewer,"span",null,null,null);\n // give HTML source to browser's parser (builds DOM nodes)\n newSpan.innerHTML=HTMLText;\n newSpan.normalize();\n // walk resulting node tree and call wikify() on each text node\n wikifyTextNodes(newSpan,highlightText,highlightCaseSensitive);\n }\n if (endHTML!=-1) // wikify everything after HTML tag\n coreWikify(tiddlerText.substr(endHTML+8),theViewer,highlightText,highlightCaseSensitive);\n // DEBUG showNodeTree(theViewer.parentNode,theViewer);\n\n}\n\nfunction wikifyTextNodes(theNode,highlightText,highlightCaseSensitive)\n{\n // pre-order traversal\n for (var i=0;i<theNode.childNodes.length;i++)\n {\n var theChild=theNode.childNodes.item(i);\n wikifyTextNodes(theChild,highlightText,highlightCaseSensitive);\n if (theChild.nodeName=='#text')\n {\n // don't bother to wikify pure whitespace nodes (if any)\n if (theChild.nodeValue.replace(/\ss+/,"").replace(/\st+/,"").length!=0)\n {\n // DEBUG alert('wikify text: "'+theChild.nodeValue.replace(regexpBackSlashEn,'\sn')+'"');\n var theClass = theNode.id.substr(0,6)=="viewer"?"viewer":theNode.id;\n var newNode = createTiddlyElement(null,"span",null,theClass,null);\n // decode newlines and macro brackets for wikification\n var theText = theChild.nodeValue.replace(regexpBackSlashEn,'\sn').replace(/\s%macro\s(/g,'<<').replace(/\s)\s%/g,'>>');\n coreWikify(theText,newNode,highlightText,highlightCaseSensitive);\n theNode.replaceChild(newNode,theChild);\n }\n }\n }\n}\n\n// Use this function to generate a report of the DOM tree objects starting from a given node.\n// place = where to display DOM object report, theNode = root of DOM object tree to be reported\nfunction showNodeTree(place,theNode)\n{\n createTiddlyElement(place,"HR",null,null,null);\n var theReport = createTiddlyElement(place,"div",null,null,null);\n walkNodeTree(theReport,theNode,'');\n}\nfunction walkNodeTree(theOutput,theNode,thePrefix)\n{\n var msg=thePrefix+':'+((theNode.nodeName=='#text')?' ':theNode.nodeName);\n var href = (theNode.href)?' href='+theNode.href:'';\n var id = (theNode.id)?' id='+theNode.id:'';\n var val = (theNode.name)?' '+theNode.name+((theNode.value)?'='+theNode.value:''):'';\n var text = (theNode.nodeName=='#text')?'"'+theNode.nodeValue.replace(regexpBackSlashEn,'\sn')+'"':'';\n if ( (theNode.nodeName!='B')\n &&(theNode.nodeName!='I')\n &&(theNode.nodeName!='TBODY')\n &&(theNode.nodeName!='SPAN'))\n createTiddlyElement(theOutput,"div",null,null,msg+val+id+href+text);\n for (var i=0;i<theNode.childNodes.length;i++)\n {\n var theChild=theNode.childNodes.item(i);\n var childmsg=msg;\n if (theNode.childNodes.length>1) childmsg+='['+(i+1)+']';\n walkNodeTree(theOutput,theChild,childmsg);\n }\n}\n
[[Introduction]] [[Research]] [[Bio]] [[Contact]] [[Selected Papers|selected recent papers]] [[Teaching]] [[Pictures]] [[Useful Links]] [[Group]][[Software]]
@media screen\n{\n\nbody {\n font: 14px/125% "Lucida Grande", "Trebuchet MS", "Bitstream Vera Sans", Verdana, Helvetica, sans-serif;\n}\n\n}\n\n@media print\n{\n\nbody {\n font-size: 6.2pt;\n font-family: "Lucida Grande", "Bitstream Vera Sans", Helvetica, Verdana, Arial, sans-serif;\n}\n\n}\n\n.viewer {\n text-align: justify;\n}\n\n
[[Programming]]\n[[Research Links]]\n[[Conference Stuff]]\n[[LaTeX and friends]]\n
<html>\n<iframe src="http://www.google.com/calendar/embed?src=csviks%40gmail.com&height=614" style=" border-width:0 " width="640" frameborder="0" height="614"></iframe> \n</html>
[[Introduction]]\n[[Research]]\n[[Bio]]\n[[Contact]]
TBA
''Abstract''\n\nWe propose an optimization algorithm to solve the Brachytherapy Seed Localization problem in prostate brachytherapy. Our algorithm is based on novel geometric approaches to exploit the special structure of the problem and relies on a number of key observations which help us formulate the optimization problem as a minimization Integer Program. Our IP model precisely defines the feasibility polyhedron for this problem, the solution to its corresponding linear program is rounded to yield an integral solution to the problem of determining correspondences between seeds in multiple projection images. The algorithm is efficient in theory as well as in practice and performs well on simulation data (98% accuracy) and real X-ray images (95% accuracy). We present in detail the underlying ideas and an extensive set of performance evaluations based on our implementation. \n\n[[PDF|http://www.biostat.wisc.edu/~vsingh/brachy.pdf]]\n\n[[bibtex|http://www.biostat.wisc.edu/~vsingh/vikas07_tmi_bib.txt]]\n\n''[[Copyright]]''
''Abstract''\n\nRecent research in biology has indicated correlations between the movement patterns of functional sites (such as replication sites in DNA) and zones of genetic activity within a nucleus. A detailed study and analysis of the motion dynamics of these sites can reveal an interesting insight into their role in DNA replication and function. In this paper, we propose a suite of novel techniques to determine, analyze, and interpret the mobility patterns of functional sites. Our algorithms are based on interesting ideas from theoretical computer science and learning and provide for the first time the tools to interpret the seemingly stochastic motion patterns of the functional sites within the nucleus in terms of a set of tractable `patterns' which can then be analyzed to understand their biological significance.\n\n[[PDF|http://www.biostat.wisc.edu/~vsingh/mob.pdf]]\n\n[[bibtex|http://www.biostat.wisc.edu/~vsingh/lopa06_miccai_bib.txt]]\n\n''[[Copyright]]''
''Abstract''\n\n[[PDF]]\n\n[[bibtex|http://www.biostat.wisc.edu/~vsingh/xu05_joco_bib.txt]]\n\n''[[Copyright]]''
''Abstract''\n\nBiplane angiographic imaging is a primary method for visual and quantitative assessment of the vasculature. In order to reliably reconstruct the three dimensional (3D) position, orientation, and shape of the vessel structure, a key problem is to determine the rotation matrix ''R'' and the translation vector ''t'' which relate the two coordinate systems. This so-called Imaging Geometry Determination problem is well studied in the medical imaging and computer vision communities and a number of interesting approaches have been reported. Each such technique determines a solution which yields 3D vasculature reconstructions with errors comparable to other techniques. From the literature, we see that different techniques with different optimization strategies yield reconstructions with equivalent errors. We have investigated this behavior, and it appears that the error in the input data leads to this equivalence effectively yielding what we call the solution space of feasible geometries, i.e., geometries which could be solutions given the error or uncertainty in the input image data. In this paper, we lay the theoretical framework for this concept of a solution space of feasible geometries using simple schematic constructions, deriving the underlying mathematical relationships, presenting implementation details, and discussing implications and applications of the proposed idea. Because the solution space of feasible geometries encompasses equivalent solutions given the input error, the solution space approach can be used to evaluate the precision of calculated geometries or 3D data based on known or estimated uncertainties in the input image data. We also use the solution space approach to calculate an imaging geometry, i.e., a solution.\n\n[[PDF|http://www.biostat.wisc.edu/~vsingh/medphys.pdf]]\n\n[[bibtex|http://www.biostat.wisc.edu/~vsingh/vikas06_medphysics_bib.txt]]\n\n''[[Copyright]]''
''Abstract''\n\nWe study the problem of classifying an autistic group from controls using structural image data alone, a task that requires a clinical interview with a psychologist. Because of the highly convoluted brain surface topology, feature extraction poses the first obstacle. A clinically relevant measure called the cortical thickness has shown promise but yields a rather challenging learning problem -- where the dimensionality of the distribution is extremely large and the training set is small. By observing that each point on the brain cortical surface may be treated as a 'hypothesis', we propose a new algorithm for LPBoosting (with truncated neighborhoods) for this problem. In addition to learning a high quality classifier, our model incorporates topological priors into the classification framework directly -- that two neighboring points on the cortical surface (hypothesis pairs) must have similar discriminative qualities. As a result, we obtain not just a label {+1,-1} for test items, but also an indication of the 'discriminative regions' on the cortical surface. We discuss the formulation and present interesting experimental results.\n\n[[PDF|http://www.biostat.wisc.edu/~vsingh/miccai08.pdf]]\n\n[[bibtex|http://www.biostat.wisc.edu/~vsingh/vikas08_miccai_bib.txt]]\n\n[[Poster|http://www.biostat.wisc.edu/~vsingh/vikasMiccai08Poster.pdf]]\n\n''[[Copyright]]''
Copyright and all rights therein are retained by authors or by other copyright holders (e.g., IEEE, ACM or other publishers).\n\nThe documents distributed by this server have been provided by the contributing authors as a means to ensure timely dissemination of scholarly and technical work on a noncommercial basis. Copyright and all rights therein are maintained by the authors or by other copyright holders, notwithstanding that they have offered their works here electronically. It is understood that all persons copying this information will adhere to the terms and constraints invoked by each author's copyright. These works may not be reposted without the explicit permission of the copyright holder.
''Abstract''\n\n[[PDF|http://www.springerlink.com/content/6511046m20100502/]]\n\n[[bibtex|http://www.biostat.wisc.edu/~vsingh/joco08.txt]]\n\n''[[Copyright]]''
[[ACM Events|http://campus.acm.org/calendar/]]\n[[Medical Imaging Conferences|http://www.isi.uu.nl/Conferences/]]\n[[Conference listings|http://www.cs.ualberta.ca/~zaiane/htmldocs/ConfRanking.html]]\n[[AI Conference rankings|http://www.cs.albany.edu/~ashwin/Conf_rank.html]]\n[[Journal rankings|http://www.eigenfactor.org/]]
''Abstract''\n\nWe study the cosegmentation problem where the objective is to segment the same object (i.e., region) from a pair of images. The segmentation for each image can be cast using a partitioning/segmentation function with an additional constraint that seeks to make the histograms of the segmented regions (based on intensity and texture features) similar. Using Markov Random Field (MRF) energy terms for the simultaneous segmentation of the images together with histogram consistency requirements using the squared \sell_2 (rather than \sell_1 ) distance, after linearization and adjustments, yields an optimization model with some interesting combinatorial properties. We discuss these properties which are closely related to certain relaxation strategies recently introduced in computer vision. Finally, we show some experimental results of the proposed approach.\n\n[[PDF|http://www.biostat.wisc.edu/~vsingh/coseg.pdf]]\n\n[[bibtex|http://www.biostat.wisc.edu/~vsingh/mukherjeeCvpr09.bib]]\n\n[[Oral Presentation Slides|http://www.biostat.wisc.edu/~vsingh/cosegtalkdist.pdf]] (latex sources also available by request)\n\nCode (coming soon), Data (available, send email)\n\n''[[Copyright]]''
''Abstract''\n\n Graph-cuts based algorithms are effective for a variety of segmentation tasks in computer vision. Ongoing research is focused toward making the algorithms even more general, as well as to better understand their behavior with respect to issues such as the choice of the weighting function and sensitivity to placement of seeds. In this paper, we investigate in the context of neuroimaging segmentation, the sensitivity/stability of the solution with respect to the input "labels" or seeds. In particular, as a form of parameter learning, we are interested in the effect of allowing the given set of labels (and consequently, the response/statistics of the weighting function) to vary for obtaining lower energy segmentation solutions. This perturbation leads to a refined label set (or parameters) better suited to the input image, yielding segmentations that are less sensitive to the set of labels or seeds provided. Our proposed algorithm (using Parametric Pseudoflow) yields improvements over graph-cuts based segmentation with a fixed set of labels. We present experiments on about 450 3-D brain image volumes demonstrating the efficacy of the algorithm.\n\n[[PDF|http://www.biostat.wisc.edu/~vsingh/vMRF.pdf]]\n\n[[bibtex|http://www.biostat.wisc.edu/~vsingh/howerIccv09.bib]]\n\n[[Poster|http://www.biostat.wisc.edu/~vsingh/vmrfPoster2.pdf]]\n\n[[Code (for the underlying parametric max-flow)|http://riot.ieor.berkeley.edu/riot/Applications/Pseudoflow/]] GUI based code link\n\n''[[Copyright]]''
''Abstract''\n\nAbstract. We present a novel computational framework for characterizing signal in brain images via nonlinear pairing of critical values of the signal. Among the astronomically large number of different pairings possible, we show that representations derived from specific pairing schemes provide concise representations of the image. This procedure yields a min-max diagram of the image data. The representation turns out to be especially powerful in discriminating image scans obtained from different clinical populations, and directly opens the door to applications in a variety of learning and inference problems in biomedical imaging. It is noticed that this strategy significantly departs from the standard image analysis paradigm – where the "mean" signal is used to characterize an ensemble of images. This offers robustness to noise in subsequent statistical analyses, for example; however, the attenuation of the signal content due to averaging makes it rather difficult to identify subtle variations. The proposed topologically oriented method seeks to address these limitations by characterizing and encoding topological features or attributes of the image. As an application, we have used this method to characterize cortical thickness measures along brain surfaces in classifying autistic subjects. Our promising experimental results provide evidence of the power of this representation.\n\n[[PDF|http://www.biostat.wisc.edu/~vsingh/pd.pdf]]\n\n[[bibtex|http://www.biostat.wisc.edu/~vsingh/pd_bib.txt]]\n\n[[Poster|http://www.biostat.wisc.edu/~vsingh/minmax_micca09Poster.pdf]]\n\n''[[Copyright]]''
''Abstract''\n\n[[PDF|http://www.biostat.wisc.edu/~vsingh/mklem.pdf]]\n\n[[bibtex|http://www.biostat.wisc.edu/~vsingh/hinrichsMiccai09.bib]]\n\n[[Poster|http://www.biostat.wisc.edu/~vsingh/mklemPoster.pdf]]\n\n''[[Copyright]]''
''Abstract''\n\nWe consider the ensemble clustering problem where the task is to `aggregate' multiple clustering solutions into a single consolidated clustering that maximizes the shared information among given clustering solutions. We obtain several new results for this problem. First, we note that the notion of agreement under such circumstances can be better captured using an agreement measure based on a $2D$ string encoding rather than voting strategy based methods proposed in literature. Using this generalization, we first derive a nonlinear optimization model to maximize the new agreement measure. We then show that our optimization problem can be transformed into a strict $0$-$1$ Semidefinite Program (SDP) via novel convexification techniques which can subsequently be relaxed to a polynomial time solvable SDP. Our experiments indicate improvements not only in terms of the proposed agreement measure but also the existing agreement measures based on voting strategies. We discuss evaluations on clustering and image segmentation databases. \n\n[[PDF|http://www.biostat.wisc.edu/~vsingh/ce.pdf]]\n\n[[bibtex|http://www.biostat.wisc.edu/~vsingh/vikas07_nips_bib.txt]]\n\n[[Poster|http://www.biostat.wisc.edu/~vsingh/vikasNips2007Poster.pdf]] (latex sources also available by request)\n\n[[Code|http://www.biostat.wisc.edu/~vsingh/ens_impl.tar.gz]] \n\n''[[Copyright]]''
[[C++|http://www.cplusplus.com/ref/]]\n[[STL|http://www.sgi.com/tech/stl/table_of_contents.html]]\n[[Linux Man pages|http://linux.ctyme.com/]]\n[[CGAL Manual|http://www.cgal.org/Manual/doc_html/index.html]]\n[[LEDA|http://www.algorithmic-solutions.info/leda_manual/MANUAL.html]]\n[[CLP User Manual|http://www.coin-or.org/Clp/userguide/index.html]]\n[[Subversion notes|http://svnbook.red-bean.com/]]\n[[VTK|http://www.vtk.org/doc/nightly/htm]]\n[[GLPK|http://www.gnu.org/software/glpk/]]\n[[GSL|http://www.gnu.org/software/gsl/]]\n[[CVX|http://www.stanford.edu/~boyd/cvx/]]\n[[Yalmip|http://control.ee.ethz.ch/~joloef/wiki/pmwiki.php?n=Main.HomePage]]\n[[CplexMEX|http://www.math.kth.se/~mpry/cplexmex.html]]\n[[CplexInt|http://control.ee.ethz.ch/~hybrid/cplexint.php]]\n[[SDPT3|http://www.math.nus.edu.sg/~mattohkc/sdpt3.html]]\n[[Sedumi|http://sedumi.ie.lehigh.edu/]]\n[[A list of Matlab interfaces to solvers|http://www.dii.unisi.it/~hybrid/tools/mex/downloads.html]]\n[[MTL|http://www.osl.iu.edu/research/mtl/]]\n[[STL Examples|http://www.josuttis.com/libbook/toc.html]]\n[[Wild Magic|http://www.geometrictools.com/]]\n[[MIPAV|http://mipav.cit.nih.gov/]]\n[[C code for Numerical Computation|http://cliodhna.cop.uop.edu/~hetrick/c-sources.html]] (old)\n[[Tcl/Tk|http://www.tcl.tk/man/tcl8.5/]]\n[[C++ User's journal|http://www.cuj.com/]]\n[[Eclipse CDT|http://www.eclipse.org/cdt/]] (only if you must)\n[[Advanced Bash|http://www.linuxdoc.org/LDP/abs/]]\n[[Bash Programming tutorial|http://steve-parker.org/sh/sh.shtml]]\n[[My .emacs|http://www.biostat.wisc.edu/~vsingh/dotemacs.txt]] (almost nothing is original...)
I grew up in [[Calcutta|http://en.wikipedia.org/wiki/Calcutta]], a city in the eastern part of India. I finished high school at [[Vikas Vidyalaya|http://en.wikipedia.org/wiki/Vikas_Vidyalaya]]. In case you are wondering what I look like, [[here it is|Pictures]].\n\n\n
''Abstract''\n\nThis paper proposes an new optimization framework for tomographic reconstruction of 3D volumes when only a limited number of projection views are available. The problem has several important clinical applications spanning coronary angiographic imaging, breast tomosynthesis and dental imaging. We first show that the limited view reconstruction problem can be formulated as a `constrained' version of the metric labeling problem. This lays the groundwork for a linear programming framework that brings together metric labeling classification and classical algebraic tomographic reconstruction (ART) in a unified model. If the imaged volume is known to be comprised of a finite set of attenuation coefficients, given a regular limited view reconstruction as an input, we can view it as a `denoising' task -- where voxels must be reassigned subject to maximally maintaining consistency with the input reconstruction and the objective of ART simultaneously. The approach can reliably reconstruct volumes with several multiple contrast objects as well as the simpler binary contrast case which can be solved near-optimally in practice. We present evaluations on cone bean computed tomography, it can also be readily extended to other tomographic modalities as a viable approach for limited-view tomographic reconstruction.\n\n[[PDF|http://www.biostat.wisc.edu/~vsingh/sparse_writeup.pdf]]\n\n[[bibtex|http://www.biostat.wisc.edu/~vsingh/singhMmbia07.bib]]\n\n''[[Copyright]]''
''Abstract''\n\nWe study the so-called Generalized Median graph problem where the task is to to construct a prototype (i.e., a `model') from an input set of graphs. The problem finds applications in many vision (e.g., object recognition) and learning problems where graphs are increasingly being adopted as a representation tool. Existing techniques for this problem are evolutionary search based; in this paper, we propose a polynomial time algorithm based on a linear programming formulation. We present an additional bi-level method to obtain solutions arbitrarily close to the optimal in non-polynomial time (in worst case). Within this new framework, one can optimize edit distance functions that capture similarity by considering vertex labels as well as the graph structure simultaneously. In context of our motivating application, we discuss experiments on molecular image analysis problems - the methods will provide the basis for building a topological map of all pairs of the human chromosome. \n\n[[PDF|http://www.biostat.wisc.edu/~vsingh/mediang.pdf]]\n\n[[bibtex|http://www.biostat.wisc.edu/~vsingh/mukherjeeIccv07.bib]]\n\n[[Poster|http://www.biostat.wisc.edu/~vsingh/vikasICCV2007Poster.pdf]] [[Poster (A4 size)|http://www.biostat.wisc.edu/~vsingh/mediang_poster_a4.pdf]] (latex sources also available by request)\n\n''[[Copyright]]''
The backdrop is the canal along Nyhavn in Copenhagen\n\n[img[DSC01894.JPG]]\n\nThe backdrop is kinkaku-ji temple in Kyoto\n\n[img[vikasKyoto.JPG]]
''Abstract''\n\nThis paper proposes a new discrete optimization framework for tomographic reconstruction and segmentation of CT volumes when only a few projection views are available. The problem has important clinical applications in coronary angiographic imaging. We first show that the limited view reconstruction and segmentation problem can be formulated as a 'constrained' version of the metric labeling problem. This lays the groundwork for a linear programming framework that brings metric labeling classification and classical algebraic tomographic reconstruction (ART) together in a unified model. If the imaged volume is known to be comprised of a finite set of attenuation coefficients (a realistic assumption), given a regular limited view reconstruction, we view it as a task of voxels reassignment subject to maximally maintaining consistency with the input reconstruction and the objective of ART simultaneously. The approach can reliably reconstruct (or segment) volumes with several multiple contrast objects. We present evaluations using experiments on cone beam computed tomography.\n\n[[PDF|http://www.biostat.wisc.edu/~vsingh/cviu08.pdf]]\n\n[[bibtex|http://www.biostat.wisc.edu/~vsingh/singhCviu08.bib]]\n\n''[[Copyright]]''
''Abstract''\n\n[[PDF|http://www.biostat.wisc.edu/~vsingh/fwght.pdf]]\n\n[[bibtex|http://www.biostat.wisc.edu/~vsingh/mukherjeeCvpr10.bib]]\n\nCode (coming soon), Data (available, send email)\n\n''[[Copyright]]''
I am assuming you can get to MSC (Medical Sciences Center), see http://map.wisc.edu/ or on [[Google Maps|http://bit.ly/4plrhM]] (''click the MSC marker in the map for more information'').
Welcome to the homepage of Vikas Singh.\n\nI am an Assistant Professor in [[Biostatistics & Med. Informatics|www.biostat.wisc.edu]] and [[Computer Sciences|www.cs.wisc.edu]] departments at [[University of Wisconsin-Madison|http://wisc.edu]]. \n\n
[[Dictionary of Algorithms and Data Structures|http://www.nist.gov/dads/]]\n[[Geometry in Medical Imaging|http://www.ics.uci.edu/%7Eeppstein/gina/medic.html]]\n[[Stony Brook Algorithm Repository|http://www.cs.sunysb.edu/%7Ealgorith/]]\n[[Compendium of NP Optimization Problems|http://www.nada.kth.se/~viggo/problemlist/compendium.html]]\n[[Opt online|http://www.optimization-online.org/]]\n[[Handbook of Computational Geometry|http://valis.cs.uiuc.edu/~sariel/research/CG/handbook.html]]\n[[Matrix cookbook|http://www2.imm.dtu.dk/pubdb/views/edoc_download.php/3274/pdf/imm3274.pdf]]\n
''Abstract''\n\nSpatially augmented LP Boosting is a machine learning technique with a bias to make it specially suited to 3D medical imaging problems; the algorithm tries to produce a linear classifier which corresponds to "smooth" contiguous regions in the brain. The "spatial augmentation" is achieved by penalizing certain variations between weights placed on neighboring voxels concurrently with the standard learning phase, i.e., smoothness is injected into the trade-off between L1-norm sparsity and training set clssification margin, leading to a 3-way trade-off.\n\n[[PDF|http://www.biostat.wisc.edu/~vsingh/spatial-lp.pdf]]\n\n[[bibtex|http://www.biostat.wisc.edu/~vsingh/hinrichsNeurolmage09.bib]]\n\n[[Code and project details|http://pages.cs.wisc.edu/~hinrichs/spLPBoost/]]\n\n''[[Copyright]]''
[[PSTricks|http://tug.org/PSTricks/main.cgi?file=docs]]\n[[PGF/Tikz|http://www.ctan.org/tex-archive/help/Catalogue/entries/pgf.html]]\n[[Seminar|http://www.tug.org/applications/Seminar/]]\n[[Prosper|http://prosper.sourceforge.net/]]\n[[Beamer|http://latex-beamer.sourceforge.net/]]\n[[Ipe|http://tclab.kaist.ac.kr/ipe/]]\n[[RefTex|http://www.gnu.org/software/auctex/reftex.html]]\n[[Bibtex for Pubmed|http://www.bioinformatics.org/texmed/]]\n[[UW-Biostat & Med. Info LaTeX letters|http://www.biostat.wisc.edu/~vsingh/vikasletter.tar.gz]] (very crude, feel free to hack)\n[[NIH Biosketch|http://www.cs.duke.edu/brd/NIH/tips/src/]]
''Abstract''\n\nabstract goes here\n\n[[PDF|http://www.biostat.wisc.edu/~vsingh/pcoseg.pdf]]\n\n[[Longer version|http://www.biostat.wisc.edu/~vsingh/pcosegl.pdf]]\n\n[[bibtex|http://www.biostat.wisc.edu/~vsingh/motwaniNips10.bib]]\n\n[[Poster|http://www.biostat.wisc.edu/~vsingh/pcosegPoster.pdf]]\n\n[[Slides|http://www.biostat.wisc.edu/~vsingh/pcosegTalk.pdf]]\n\nCode and project details\n\n''[[Copyright]]''
Look under the details section of the corresponding paper.
''Abstract''\n\nWe study the problem of segmenting specific white matter structures of interest from Diffusion Tensor (DTMR) images of the human brain. This is an important requirement in many Neuroimaging studies: for instance, to evaluate whether a brain structure exhibits group level differences as a function of disease in a set of images. Typically, interactive expert guided segmentation has been the method of choice for such applications, but this is tedious for larger datasets common today (> 200 images). To address this problem, the strategy we adopt is to endow an image segmentation algorithm with "advice" encoding some global characteristics of the region(s) we want to extract. This is accomplished by constructing (using expert-segmented images) an epitome of a specific region - as a histogram over a bag of 'words' (e.g., suitable feature descriptors). Now, given such a representation, the problem reduces to segmenting a new 3-D DTMR image of the brain with additional constraints that enforce consistency between the segmented foreground and the pre-specified histogram (over features). We present combinatorial approximation algorithms to incorporate such domain specific constraints for Markov Random Field (MRF) segmentation. Making use of recent results on concurrent segmentation of multiple images, we derive effective solution strategies for our problem. We describe our main ideas, provide analysis of solution quality, and present promising experimental evidence showing that many structures of interest in Neuroscience can be extracted reliably from 3-D DTMR brain image volumes using our algorithm.\n\n[[PDF|http://www.biostat.wisc.edu/~vsingh/pcoseg.pdf]]\n\n[[Longer version|http://www.biostat.wisc.edu/~vsingh/pcosegl.pdf]]\n\n[[bibtex|http://www.biostat.wisc.edu/~vsingh/motwaniNips10.bib]]\n\n[[Poster|http://www.biostat.wisc.edu/~vsingh/pcosegPoster.pdf]]\n\n[[Slides|http://www.biostat.wisc.edu/~vsingh/pcosegTalk.pdf]]\n\n[[Code and project details|http://pages.cs.wisc.edu/~kmotwani/nips10/]]\n\n''[[Copyright]]''
''Abstract''\n\nAlzheimer’s Disease (AD) and other neurodegenerative diseases affect over 20 million people worldwide, and this number is projected to significantly increase in the coming decades. Proposed imaging based markers have shown steadily improving levels of sensitivity/specificity in classifying individual subjects as AD or normal. Several of these efforts have utilized statistical machine learning techniques, using brain images as input, as means of deriving such AD related markers. A common characteristic of this line of research is a focus on either (1) using a single imaging modality for classification, or (2) incorporating several modalities, but reporting separate results for each. One strategy to improve on the success of these methods is to leverage all available imaging modalities together in a single automated learning framework. The rationale is that some subjects may show signs of pathology in one modality but not in another – by combining all available images a clearer view of the progression of disease pathology will emerge. Our method is based on the Multi Kernel Learning (MKL) framework, which allows the inclusion of an arbitrary number of views of the data in a maximum margin, kernel learning framework. The principal innovation behind MKL is that it learns an optimal combination of kernel (similarity) matrices while simultaneously training a classifier. In classification experiments MKL outperformed an SVM trained on all available features by 3% – 4%. We are especially interested in whether such markers are capable of identifying early signs of the disease. To address this question, we have examined whether our multi-modal disease marker (MMDM) can predict conversion from Mild Cognitive Impairment (MCI) to AD. Our experiments reveal that this measure shows significant group differences between MCI subjects who progressed to AD, and those who remained stable for 3 years. These differences were most significant in MMDM based on imaging data. We also discuss the relationship between our MMDM and an individual’s conversion from MCI to AD.\n\n[[PDF|http://www.biostat.wisc.edu/~vsingh/mkl-adni.pdf]]\n\n[[bibtex|http://www.biostat.wisc.edu/~vsingh/hinrichsNeurolmage10.bib]]\n\n[[Code and project details|http://pages.cs.wisc.edu/~hinrichs/MKL_ADNI/]]\n\n''[[Copyright]]''
''Abstract''\n\nDiffusion Tensor Imaging (DTI) provides estimates of local directional information regarding paths of white matter tracts in the human brain. An important problem in DTI is to infer tract connectivity (and networks) from given image data. We propose a method that infers high-level network structures and connectivity information from Diffusion Tensor images. Our algorithm extends principles from perceptual contours to construct a weighted line-graph based on how well the tensors agree with a set of proposal curves (regularized by length and curvature). The problem of extracting high-level anatomical connectivity is then posed as an optimization problem over this curvature-regularizing graph — which gives subgraphs which comprise a representation of the tracts' network topology. We present experimental results and an open-source implementation of the algorithm.\n\n[[PDF|http://www.biostat.wisc.edu/~vsingh/lgopt.pdf]]\n\n[[Longer version|http://www.biostat.wisc.edu/~vsingh/lgoptl.pdf]]\n\n[[bibtex|http://www.biostat.wisc.edu/~vsingh/collinsAccv10.bib]]\n\n[[Slides|http://www.biostat.wisc.edu/~vsingh/collinsAccvTalk.pdf]]\n\nCode and project details\n\n''[[Copyright]]''
''Abstract''\n\nIn this paper, we study the ensemble clustering problem, where the input is in the form of multiple clustering solutions. The goal of ensemble clustering algorithms is to aggregate the solutions into one solution that maximizes the agreement in the input ensemble. We obtain several new results for this problem. Specifically, we show that the notion of agreement under such circumstances can be better captured using a 2D string encoding rather than a voting strategy, which is common among existing approaches. Our optimization proceeds by first constructing a non-linear objective function which is then transformed into a 0-1 Semidefinite program (SDP) using novel convexification techniques. This model can be subsequently relaxed to a polynomial time solvable SDP. In addition to the theoretical contributions, our experimental results on standard machine learning and synthetic datasets show that this approach leads to improvements not only in terms of the proposed agreement measure but also the existing agreement measures based on voting strategies. In addition, we identify several new application scenarios for this problem. These include combining multiple image segmentations and generating tissue maps from multiple-channel Diffusion Tensor brain images to identify the underlying structure of the brain.\n\n[[PDF|http://www.biostat.wisc.edu/~vsingh/ceml.pdf]]\n\n[[bibtex|http://www.biostat.wisc.edu/~vsingh/singhML10.bib]]\n\n[[Code|http://www.biostat.wisc.edu/~vsingh/ens_impl.tar.gz]] \n\n''[[Copyright]]''
''Abstract''\n\nMaximal margin based frameworks have emerged as a powerful tool for supervised learning. The extension of these ideas to the unsupervised case, however, is problematic since the underlying optimization entails a discrete component. In this paper, we first study the computational complexity of maximal hard margin clustering and show that the hard margin clustering problem can be precisely solved in $O(n^{d+2})$ time where $n$ is the number of the data points and $d$ is the dimensionality of the input data. However, since it is well known that many datasets commonly express themselves primarily in far fewer dimensions, our interest is in evaluating if a careful use of dimensionality reduction can lead to practical and effective algorithms. We build upon these observations and propose a new algorithm that gradually increases the number of features used in the separation model in each iteration, and analyze the convergence properties of this scheme. We report on promising numerical experiments based on a ‘truncated’ version of this approach. Our experiments indicate that for a variety of datasets, good solutions equivalent to those from other existing techniques can be obtained in significantly less time.\n\n[[PDF|http://www.biostat.wisc.edu/~vsingh/mm.pdf]]\n\n[[bibtex|http://www.biostat.wisc.edu/~vsingh/pengJGO11.bib]]\n\n''[[Copyright]]''
I received a Ph.D in 2007 in [[Computer Science|http://www.cse.buffalo.edu/]] at [[State University of New York at Buffalo|http://www.buffalo.edu/]]. There, I was affiliated with [[Toshiba Stroke Research Center|http://www.ubneurosurgery.com/handler.cfm?event=practice,template&cpid=6383]]. Along the way, I received a Masters in 2004. Earlier, I finished my undergraduate degree in Computer Science and Engineering in 2002 from [[Uttar Pradesh Technical University|http://www.uptu.ac.in]].\n\nIf you are looking for additional information on me, go [[here|Personal]]. \n\n
''Abstract''\n\nOur primary interest is in generalizing the problem of Cosegmentation to a large group of images, that is, concurrent segmentation of common foreground region(s) from multiple images. We further wish for our algorithm to offer scale invariance (foregrounds may have arbitrary sizes in different images) and the running time to increase (no more than) near linearly in the number of images in the set. What makes this setting particularly challenging is that even if we ignore the scale invariance desiderata, the Cosegmentation problem, as formalized in many recent papers (except [1]), is already hard to solve optimally in the two image case. A straightforward extension of such models to multiple images leads to loose relaxations; and unless we impose a distributional assumption on the appearance model, existing mechanisms for image-pair-wise measurement of foreground appearance variations lead to significantly large problem sizes (even for moderate number of images). This paper presents a surprisingly easy to implement algorithm which performs well, and satisfies all requirements listed above (scale invariance, low computational requirements, and viability for the multiple image setting). We present qualitative and technical analysis of the properties of this framework.\n\n[[PDF|http://www.biostat.wisc.edu/~vsingh/mcoseg.pdf]]\n\n[[Longer version|http://www.biostat.wisc.edu/~vsingh/mcosegl.pdf]]\n\n[[bibtex|http://www.biostat.wisc.edu/~vsingh/mukherjeeCvpr11.bib]]\n\n[[Slides|http://www.biostat.wisc.edu/~vsingh/mcosegTalk.pdf]]\n\nCode and project details\n\n''[[Copyright]]''
''Abstract''\n\nWe propose a new algorithm for learning kernels for variants of the Normalized Cuts (N-cuts) objective, i.e., given a set of training examples with known partitions, how should a basis set of similarity functions be combined to induce N-cuts favorable distributions. Such a procedure facilitates design of good affinity matrices. It also helps assess the importance of different feature types for discrimination. Rather than formulating the learning problem in terms of the spectral relaxation, the alternative we pursue here is to work in the original discrete setting (i.e., the relaxation occurs much later). We show that this strategy is useful: while the initial specification seems rather difficult to optimize efficiently, a set of manipulations reveal a related model which permits a nice SDP relaxation. A salient feature of our model is that the eventual problem size is only a function of the number of input kernels and not the training set size. This relaxation also allows strong optimality guarantees, if certain conditions are satisfied. We show that the sub-kernel weights obtained provide a complementary approach for MKL based methods. Our experiments on Caltech101 and ADNI (a brain imaging dataset) show that the quality of solutions is competitive with the state-of-the-art.\n\n[[PDF|http://www.biostat.wisc.edu/~vsingh/fwght.pdf]]\n\n[[Longer version|http://www.biostat.wisc.edu/~vsingh/fwghtl.pdf]]\n\n[[bibtex|http://www.biostat.wisc.edu/~vsingh/mukherjeeCvpr10.bib]]\n\n[[Poster|http://www.biostat.wisc.edu/~vsingh/fwghtPoster.pdf]]\n\n[[Code and project details|http://pages.cs.wisc.edu/~hinrichs/feature_weighting/]]\n\n[[Talk|http://videolectures.net/cvpr2010_spotlights10/]]\n\n''[[Copyright]]''
My research is in ''Computer Vision'', ''Biomedical Image Analysis'', and some aspects of ''Machine Learning''. \nIn particular, I am interested in problems motivated from image data with a distinct optimization and/or geometric flavor.\n\nPlease take a look at some [[selected recent papers]] to get a sense of what ''[[our group|#Group]]'' is working on. \n\nIf you are a ''CS'' graduate student here interested in topics related to my research, send me email or drop by (see [[directions|Directions to MSC 5795]]). Ideally, you will have prior experience in one or more of the following areas:\n#Algorithms for Image Segmentation, Registration,...\n#Kernel methods in Computer Vision\n#Applied aspects of Convex optimization or Combinatorial methods\n#Applications of Computational Geometry to Computer Vision & Image analysis\n#Neuroscience and/or Neuroimaging\n\n@@color(red):''Undergraduate Research (for UW students only):''@@ Send me email, some positions may become available. \n\nI do ''NOT'' offer summer internships. Please do not contact me regarding this.
''Spring 2012'': I am teaching ''[[CS 766: Computer Vision|http://vsingh-www.cs.wisc.edu/cs766-12/]]''. \nWhere: 2535 ENGR HALL.\nWhen: 11:00AM--12:15PM Tues/Thurs.\n\nIn the past...\nSpring 2011: [[CS 540: Intro to AI|http://vsingh-www.cs.wisc.edu/cs540-11/]]. \nSpring 2010: [[CS 540: Intro to AI|http://vsingh-www.cs.wisc.edu/cs540-10/]].\nSpring 2009: [[CS 638: Methods in Medical Image Anal|http://www.biostat.wisc.edu/~vsingh/cs638Spring2009.html]].\nSpring 2008: [[CS 638: Methods in Medical Image Anal|http://www.biostat.wisc.edu/~vsingh/cs638Spring2008.html]]. \n\n[[Other AI courses offered here at Wisconsin|http://www.cs.wisc.edu/areas/ai]]\n
''Abstract''\n\nAlzheimer's disease (AD) research has recently witnessed a great deal of activity focused on developing new statistical learning tools for automated inference using imaging data. The workhorse for many of these techniques is the Support Vector Machine (SVM) framework (or more generally kernel based methods). Most of these require, as a first step, specification of a kernel matrix K between input examples (i.e., images). The inner product between images Ii and Ij in a feature space can generally be written in closed form, and so it is convenient to treat K as "given". However, in certain neuroimaging applications such an assumption becomes problematic. As an example, it is rather challenging to provide a scalar measure of similarity between two instances of highly attributed data such as cortical thickness measures on cortical surfaces. Note that cortical thickness is known to be discriminative for neurological disorders, so lever- aging such information in an inference framework, especially within a multi-modal method, is potentially advantageous. But despite being clinically meaningful, relatively few works have successfully exploited this measure for classification or regression. Motivated by these applications, our paper presents novel techniques to compute similarity matrices for such topologically- based attributed data. Our ideas leverage recent developments to characterize signals (e.g., cortical thickness) motivated by the persistence of their topological features, leading to a scheme for simple constructions of kernel matrices. As a proof of principle, on a dataset of 356 subjects from the ADNI study, we report good performance on several statistical inference tasks without any feature selection, dimensionality reduction, or parameter tuning. \n\n[[PDF|http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3245735/pdf/nihms344167.pdf]]\n\n[[bibtex|http://www.biostat.wisc.edu/~vsingh/pachauriTmi11.bib]]\n\n''[[Copyright]]''
Office:\n\n''[[5795 Medical Sciences Center|Directions to MSC 5795]]''\n1300 University Avenue\nMadison, WI 53706.\nemail: userid: vsingh, domain: biostat wisc edu\n\nPhone: (608)262-8875. @@color(red):(do NOT leave a message here, instead send email)@@\nFax: (608)265-7916\n\n \n\n\n\n
''Abstract''\n\nThis paper is focused on the Co-segmentation problem -- where the objective is to segment a similar object from a pair of images. The background in the two images may be arbitrary; therefore, simultaneous segmentation of both images must be performed with a requirement that the appearance of the two sets of foreground pixels in the respective images are consistent. Existing approaches [1, 2] cast this problem as a Markov Random Field (MRF) based segmentation of the image pair with a regularized difference of the two histograms -- assuming a Gaussian prior on the foreground appearance or by calculating the sum of squared differences. Both are interesting formulations but lead to difficult optimization problems, due to the presence of the second (histogram difference) term. The model proposed here bypasses measurement of the histogram differences in a direct fashion; we show that this enables obtaining efficient solutions to the underlying optimization model. Our new algorithm is similar to the existing methods in spirit, but differs substantially in that it can be solved to optimality in polynomial time using a maximum flow procedure on an appropriately constructed graph. We discuss our ideas and present promising experimental results.\n\n[[PDF|http://www.biostat.wisc.edu/~vsingh/co-seg.pdf]]\n\n[[bibtex|http://www.biostat.wisc.edu/~vsingh/xxx.txt]]\n\n[[Oral Presentation Slides|http://www.biostat.wisc.edu/~vsingh/co-segtalkdist.pdf]] \n\n[[Code|http://www.biostat.wisc.edu/~vsingh/restricted/myfiles/coseg_vikas_maxflow_distrib.tar.gz]] [[Data|http://www.cs.wisc.edu/~vsingh/pairimages.tar.gz]]\n\n@@color(blue):''The code is heavily based on [[Hochbaum's Pseudoflow|http://riot.ieor.berkeley.edu/Applications/Pseudoflow/maxflow.html]]. If you make use of this distribution, we request that you acknowledge and refer the Pseudoflow paper also (Operations Research, Volume 58(4), 992-1009, 2008).'' @@\n\n''[[Copyright]]''
''[[Complete List|http://www.biostat.wisc.edu/~vsingh/researchexhibit.html]]''\n\n> Deepti Pachauri, Chris Hinrichs, Moo K. Chung, Sterling C. Johnson, Vikas Singh, "[[Topology based Kernels with Application to inference problems in Alzheimer's disease]]", IEEE Transactions on Medical Imaging (TMI), Volume 30(10), 2011.\n\n> Lopamudra Mukherjee, Vikas Singh, Jiming Peng, "[[Scale Invariant cosegmentation for image groups]]", Proceedings of Conference on Computer Vision and Pattern Recognition (CVPR), June 2011. [also selected for ''Oral'' presentation, acceptance 3.5%]\n\n> Jiming Peng, Lopamudra Mukherjee, Vikas Singh, Dale Schuurmans, Linli Xu, "[[An efficient algorithm for maximal margin clustering]]", Journal of Global Optimization, Volume 52(1), 2011. \n\n> Chris Hinrichs, Vikas Singh, Guofan Xu, Sterling C. Johnson, "[[Predictive Markers for AD in a Multi-Modality Framework: An Analysis of MCI Progression in the ADNI Population]]", Neuroimage, Volume 55(2), March 2011. [impact factor: 5.74].\n\n> Kamiya Motwani, Nagesh Adluru, Chris Hinrichs, Andrew L. Alexander, Vikas Singh, "[[Epitome driven 3-D Diffusion Tensor image segmentation: on extracting specific structures]]", Advances in Neural Information Processing Systems (NIPS) 24, December 2010 [also selected for ''Oral spotlight'' 5.9%, Motwani received NIPS 2010 travel award, and additional funds from [[Women in Machine Learning (WIML)|http://www.wimlworkshop.org/]]].\n\n> Maxwell D. Collins, Vikas Singh, Andrew L. Alexander, "[[Network Connectivity inference over curvature regularizing line graphs]]", Proceedings of Asian Conference on Computer Vision (ACCV), November 2010. [also selected for ''Oral'' presentation, acceptance 4.7%, ''Best Application Paper'' award].\n\n> Lopamudra Mukherjee, Vikas Singh, Jiming Peng, Chris Hinrichs, "[[Learning Kernels for variants of Normalized Cuts: Convex Relaxations and Applications]]", Proceedings of Conference on Computer Vision and Pattern Recognition (CVPR), June 2010. [acceptance 27%].\n\n> Vikas Singh, Lopamudra Mukherjee, Jiming Peng, Jinhui Xu, "[[Ensemble Clustering using Semidefinite Programming with applications]]", Machine Learning, Volume 79(1-2), May 2010. [impact factor: 2.326, longer version of NIPS 2007 paper].\n\n> Dorit S. Hochbaum, Vikas Singh, "[[An efficient algorithm for co-segmentation]]", Proceedings of International Conference on Computer Vision (ICCV), October 2009. [also selected for ''Oral'' presentation, acceptance 3.6%].\n\n> Dylan Hower, Vikas Singh, Sterling C. Johnson, "[[Label Set Perturbation for MRF based Neuroimaging Segmentation]]", Proceedings of International Conference on Computer Vision (ICCV), October 2009. [acceptance 19.6%].\n\n> Chris Hinrichs, Vikas Singh, Guofan Xu, Sterling C. Johnson, "[[MKL for robust multi-modal AD classification]]", Proceedings of Medical Image Computing and Computer Assisted Intervention (MICCAI), September 2009. [acceptance 32%].\n\n> Moo K. Chung, Vikas Singh, Peter T. Kim, Kim M. Dalton, Richard J. Davidson, "[[Topological characterization of signal in brain images using the min-max diagram]]", Proceedings of Medical Image Computing and Computer Assisted Intervention (MICCAI), September 2009. [acceptance 32%].\n\n> Chris Hinrichs, Vikas Singh, Lopamudra Mukherjee, Guofan Xu, Moo K. Chung, Sterling C. Johnson, "[[Spatially augmented LP Boosting for AD classification with evaluations on the ADNI dataset]]", Neuroimage, Volume 48(1), October 2009. [impact factor: 5.74].\n\n> Lopamudra Mukherjee, Vikas Singh, Charles R. Dyer, "[[Half-integrality based algorithms for Cosegmentation of Images]]", Proceedings of Conference on Computer Vision and Pattern Recognition (CVPR), June 2009. [also selected for ''Oral'' presentation, acceptance 4.1%].\n\n> Lopamudra Mukherjee, Vikas Singh, Jiming Peng, Jinhui Xu, Michael J. Zeitz, Ronald Berezney, "[[Generalized Median Graphs and Applications]]", Journal of Combinatorial Optimization (JCO), Volume 17(1), January 2009. [Longer version of ICCV 2007 paper].\n\n> Vikas Singh, Lopamudra Mukherjee, Petru M. Dinu, Jinhui Xu, Kenneth R. Hoffmann, "[[Limited view CT reconstruction and segmentation via constrained metric labeling]]", Computer Vision and Image Understanding (CVIU), //Special Issue on Discrete Optimization in Computer Vision//, Volume 112(1), October 2008. [Longer version of MMBIA 2007 paper ''invited'' to International Journal of Computer Vision].\n\n> Vikas Singh, Lopamudra Mukherjee, Moo K. Chung, "[[Cortical surface thickness as a classifier: Boosting for autism classification]]", Proceedings of Medical Image Computing and Computer Assisted Intervention (MICCAI), September 2008. \n\n> Vikas Singh, Lopamudra Mukherjee, Jiming Peng, Jinhui Xu, "[[Ensemble Clustering using Semidefinite Programming]]", Advances in Neural Information Processing Systems (NIPS) 20, December 2007 [also selected for ''Oral spotlight'' 10.3%]. \n\n> Lopamudra Mukherjee, Vikas Singh, Jiming Peng, Jinhui Xu, Michael J. Zeitz, Ronald Berezney, "[[Generalized Median Graphs: Theory and Applications]]", Proceedings of IEEE International Conference on Computer Vision (ICCV), October 2007. [acceptance: 23.6%].\n\n> Vikas Singh, Lopamudra Mukherjee, Jinhui Xu, Kenneth R. Hoffmann, Petru M. Dinu, Matthew Podgorsak, "[[Brachytherapy Seed Localization using Geometric and Linear Programming Techniques]]", IEEE Transactions on Medical Imaging (TMI), Special Issue on Mathematical Methods in Biomedical Image Analysis, Volume 26(9), September 2007.\n\n> Vikas Singh, Jinhui Xu, Kenneth R. Hoffmann, Guang Xu, Zhenming Chen, Anant Gopal, "[[Towards a theory of a Solution Space for the Biplane Imaging Geometry problem]]", Medical Physics, Volume 33(10), October 2006.\n\n> Lopamudra Mukherjee, Vikas Singh, Jinhui Xu, Kishore Malyavantham, Ronald Berezney, "[[On Mobility Analysis of Functional Sites from Time Lapse microscopic image sequences of living Cell Nucleus]]", Proceedings of Medical Image Computing and Computer Assisted Intervention (MICCAI), October 2006.\n\n> Jinhui Xu, Guang Xu, Zhenming Chen, Vikas Singh, Kenneth R. Hoffmann, "[[Efficient Algorithms for Determining 3-D Bi-Plane Imaging Geometry]]", Journal of Combinatorial Optimization (JCO), Volume 10(2), September 2005.\n\n\n
[[Maxwell Collins|http://pages.cs.wisc.edu/~mcollins/]] (CS Doctoral student)\n[[Chris Hinrichs|http://pages.cs.wisc.edu/~hinrichs/]] (CS Doctoral student, ABD)\n[[Vamsi Ithapu|https://sites.google.com/site/vamsikrishnaithapu/home]] (ECE Doctoral student)\n[[WonHwa Kim|http://www.cs.wisc.edu/users/wonhwa]] (CS Doctoral student)\n[[Deepti Pachauri|http://pages.cs.wisc.edu/~pachauri/]] (CS Doctoral student)\n[[Vikas Singh|http://www.biostat.wisc.edu/~vsingh/]] (Principal Investigator)\n[[Jia Xu|http://pages.cs.wisc.edu/~jiaxu]] (CS Doctoral student)\n\n\n''Alumni (and former short term visitors)''\n[[Kamiya Motwani|http://pages.cs.wisc.edu/~kmotwani/]] (MS Computer Science, 2011)\n[[Patrick Blesi|http://pages.cs.wisc.edu/~blesi/]] (BS Computer Science, 2011)\nJaime Torres (undergrad, Summer 2010 as part of Summer Institute in Biostatistics)\nDylan Hower (undergrad/grad, Summer - Fall 2008)\n\n\nAt UW, we have strong ongoing collaborations with\n[[Andy Alexander|http://www.waisman.wisc.edu/faculty/alexander.html]]\n[[Chuck Dyer|http://pages.cs.wisc.edu/~dyer/]]\n[[Moo K. Chung|http://www.stat.wisc.edu/~mchung/]]\n[[Sterling Johnson|http://www.brainmap.wisc.edu/]].\n\n\nOur research is made possible due to support from \n<html>\n<img src="warf-logo.JPG" width="350" /> <img src="SIIMLogo.JPG" width="150" /> <img src="wpp.jpg" width="180" />\n\n<img src="uw-adrc.png" width="250" /> <img src="nih_logo.JPG" width="150" /> <img src="logo_nsf_world.JPG" width="150"/>\n</html>\n\n\n