PHILIPPE BURLINA
BIO | RESEARCH | BIOMEDICAL IMAGING | VISION | PUBLICATIONS | COLLABORATORS | CONTACT
I lead the machine vision team at the Research Center of the Johns Hopkins University Applied Physics Laboratory. I am also an assistant research professor at the Johns Hopkins University Department of Computer Science and a research faculty at the Johns Hopkins University Wilmer Eye Institute. My research interests span several areas of machine perception and software engineering, including machine vision, biomedical image analysis, hyperspectral video, machine learning, enterprise software systems, content and e-process management, and SDLC.
Prior to joinning JHU, I was a V.P. of engineering at eGrail and director of software development at
FileNET (now part of IBM) where I led the development of their enterprise content management and eprocess management platforms. Prior to that, with other University of
Maryland researchers, I co-founded ImageCorp
(now part of SAIC), a company dedicated to R&D in Machine Vision. I received an M.S. and a Ph.D. in Electrical Engineering
from the University of
Maryland at College Park in Communications and Control, and a Diplôme d’Ingénieur
from the Université de Technologie
de Compiègne, France.
Biomedical
Imaging: image analysis for pre-operative surgical planning, patient
specific biomechanical modeling, real-time 3D echocardiographic image analysis,
detection of retinal pathologies, cell genealogy tracking in microfluidic microscopy imagery.
Machine Vision: tracking, motion analysis, bio-mimetic
algorithms, distributed visual sensing, machine learning for machine vision.
Hyperspectral
Video: tracking, detection, and
classification in hyperspectral image and video.
Preoperative surgical planning
Our work deals with the exploitation of real time 3D echographic imaging data
to develop accurate, patient-specific computational models of the heart to allow
surgeons to preoperatively test and simulate reconstructive cardiovascular surgical interventions.
Cardiovascular disease (CVD) is one of the leading causes of death among Americans. The primary treatment for many types of CVDs entails some form of cardiac surgical reconstruction. The complex physiology and 3D anatomy of the heart presents substantial challenges when performing many of these reconstructive operations. These repairs are generally performed on an arrested heart under cardiopulmonary bypass. Because of the complexity and critical nature of this problem, it is important to develop computer aided methods allowing surgeons to preoperatively create more precise surgical plans for a given patient.
Our principal use case is mitral valve disease because of its significance among CVDs and its clinical relevance.
4D Ultrasound Image Exploitation
Real
time 3D ultrasound (also called 4D ultrasound) is an
imaging modality that is unmatched for its ability to image the
very fast motion of certain structures in the heart complex, such as
the mitral
and aortic valves. It has other benefits when compared to other 3D
imaging modalities in terms of costs, form factor, and safety. It has
however
certain drawbacks in terms of image quality (artifacts, obscuration, resolution) which make the automatic segmentation of
anatomical structures more challenging.
Patient Specific Biomechanical
Modeling
Part
of our efforts
have consisted in developing computational and physical models of the
mitral valve and the heart complex to predict the outcome of a
candidate surgical procedure, answering the question "will the
candidate reconstructed valve close competently?”. Our modeling
approach takes as a starting point the patient specific valve anatomy,
recovered from 3D echocardiography, and modified to incorporate the
surgeon's proposed reconstruction. It then performs a shape
finding procedure using a finite element approach to predict the closed
valve configuration based on physical modeling of the closure. It
assumes that the valve is subject to several forces including blood
pressure, internal hyperelastic forces, tethering forces from the
chordae tendineae, and collision forces to avoid leaflets
intersection. We predict the closed valve configuration by using an energy
minimization process which finds the valve at the closed equilibrium position during systole.
The figure below shows a set of intermediary steps in the computation of the
closed valve configuration from an assumed open configuration. We have also been
developing dynamical models that aim to evaluate the behavior of the valve when
immersed in patient specific hemodynamic conditions.
Hemodynamic Characterization
In addition to modeling,
we have been pursuing validation and have been interested in means by which
echographic imagery can be combined with echogenic contrast agents to tease out
myocardial and blood velocity fields information. The images below show the
computation of blood flow through the intraventricular chamber during diastolic
and systolic phases using contrast enhanced ultrasound. This information can be
exploited using machine learning techniques to provide automated diagnostics.
It is also important from a scientific perspective to elucidate the mechanisms
that accompany certain heart pathologies such as hypertrophic cardiomyopathy.
Retinal Pathologies
We have been
interested in developing novel methods to pre-screen individuals for retinal
anomalies such as Age Related Macular Degeneration (AMD) or Diabetic
Retinopathy. AMD is often asymptomatic in early and intermediate stages with
regard to vision loss. When undetected and untreated, AMD can advance to a
neo-vascular form leading to blindness. We have developed methods to
automatically find individuals with intermediate stage AMD which are candidate
for anti-Vascular Endothelial Growth Factor (VEGF) therapy where vision can be
stabilized. Our diagnostic relies on automated detection of drusen in fundus imagery
using machine learning techniques.
Analysis of Cell Microscopy Video
We are interested in
methods for tracking the movement of multiple cells and their lineage in microfluidics chips. We have developed methods based on the
Gaussian Mixture Probability Hypothesis Density (GM-PHD) filter, a multi-target
tracking algorithm, to track the motion of multiple cells over time and to keep
track of the lineage of cells as they spawn.
Visual Tracking
We have been
exploring novel methods for performing visual tracking and Bayesian filtering.
We have developed kernel based methods to perform efficient Bayesian filtering
using particle filters. Particle
filters (PFs) are Bayesian filters capable of modeling non-linear,
non-Gaussian, and non-stationary dynamical systems. Recent research in PFs has investigated ways
to appropriately sample from the posterior distribution, maintain multiple
hypotheses, and alleviate computational costs while preserving tracking
accuracy. To address these issues, we
have proposed to use Support Vector Data Description (SVDD) density estimation
methods within the particle filtering framework. The SVDD density
estimate can be integrated into various types of PFs yielding several benefits. It yields a sparse
representation of the posterior density that reduces the computational
complexity of the PF. The proposed
approach also provides an analytical expression for the posterior distribution
that can be used to identify its modes for maintaining multiple hypotheses and
computing the MAP estimate, and to directly sample from the posterior.
Distributed Vision
Computer
vision research has traditionally considered that visual information
acquired by multiple cameras is gathered together and processed by a
single central computer. The alternative scenario we have been
investigating involves multiple cameras trying to reach a consensus
about what they see by locally processing partial information while
communicating with only a few of their neighboring cameras through
communication channels with limited bandwidth. We have examined
how such processing cameras can best reach a consensus about the pose
of an object when they each know a model of the object, defined by a
set of world point coordinates, but can potentially only see a subset
of these points in the midst of clutter points from the background, not
knowing at first which image points match which object points, nor
which points are object points or background points. We have shown that
the cameras can reach the most accurate pose consensus by exchanging
the parameters characterizing the object's pose which are generated by
3D world coordinates penalized to agree with the input model. The
cameras use these parameters to reconstruct the object's world
coordinates using their knowledge of the model, and perform consensus
updates on these world coordinates.
Hyperspectral Video
We
are interested in video exploitation using video rate hyperspectral
sensors. Hyperspectral video cameras are much like traditional cameras
but measure dozen or hundreds of spectral bands in place of the three
RGB channels found in current video cameras, therefore offering additional benefits for detection,
classification, or tracking tasks. The exploitation and automated
analysis of visual input has been the goal of the computer vision
community, while hyperspectral cameras have been mostly studied and
exploited by the remote sensing research community. Video rate
hyperspectral cameras offer a triplet of spatial, temporal and spectral
information and promise to combine capabilities studied by both fields
of research.
PUBLICATIONS
Journals and book chapters
B. Hoffmann and P. Burlina, “Ultrasound in Trauma – From Civilian to the Combat Casualty Environment”, Shock, 2011, in press.
H.V. Nguyen,
A. Banerjee, P. Burlina, J. Broadwater, R. Chellappa, "Tracking And Id Via
Object Reflectance Using A Hyperspectral Video Camera", Machine Vision
Beyond Visible Spectrum, Guoliang Fan (Ed.), Springer
2011.
P. Burlina and
R. Chellappa, “Temporal Analysis of Motion in Video Sequences through
Predictive Operators”, International Journal of Computer Vision, Vol. 28, No.
2, pp. 175-192, 1998.
P. Burlina and
F. Alajaji, “An Error Resilient Scheme for Image Transmission over Noisy
Channels with Memory”, IEEE Transactions on Image Processing, Vol. 7, No. 4,
pp. 593-600, April 1998.
R. Chellappa,
P. Burlina, X. Zhang, Q. Zheng, C. L. Lin, V. Parameswaran,
L. Davis, and A. Rosenfeld, “Site Model Mediated Detection of Movable Objects
Activities”, in RADIUS: Image Understanding for Imagery Intelligence, (O.Firshein, ed.), Morgan Kaufman, pp 285-317, May 1997.
R. Chellappa,
Q. Zheng, S. Kuttikkad, C. Shekhar, and P. Burlina,
“Site Model Construction for the Exploitation of EO and SAR Imagery”, in
RADIUS: Image Understanding for Imagery Intelligence, (O.Firshein,
ed.), Morgan Kaufman, pp 185-208, May 1997.
R. Chellappa,
Q. Zheng, C. Shekhar, and P. Burlina, “Site Model Supported Targeting”, in
RADIUS: Image Understanding for Imagery Intelligence, (O.Firshein,
ed.), Morgan Kaufman, pp 357-369, May 1997.
R. Chellappa,
Q. Zheng, P. Burlina, C. Shekhar, and K. B. Eom, “On
the Positioning of Multisensor Imagery for
Exploitation and Target Recognition”, Proceedings of the IEEE, Vol. 85, No. 1,
pp 120-138, January 1997.
P. Burlina and
R. Chellappa, “Analyzing Looming Motion Components from their Spatiotemporal
Spectral Signature”, IEEE Transactions on Pattern Analysis and Machine Intelligence,
Vol. 18, No. 10, pp 1029-1034, October 1996.
R. Chellappa,
T. Wu, P. Burlina, and Q. Zheng, “Visual Motion Analysis”, in Control and
Dynamic Systems: Advances in Theory and Applications, Vol. 67, (C.T. Leondes, ed.), Academic Press, 1994.
Conference and workshops
P.
Burlina, B. Hoffmann, and T. Abraham, “Computing Left Ventricular
Hemodynamics from Echographic Optical Flow of CEUS Microspheres”, in
Proc. IEEE NIH Conf. Life Science Systems and Applications, 2011.
P.
Burlina R. Mukherjee, R. Juang, C. Sprouse, “Recovering Endocardial
Walls from 3D TEE”, in Proc. Sixth Int. Conf. on Functional Imaging and
Modeling of the Heart, 2011.
R. Juang, E.R. McVeigh, B. Hoffmann, D.
Yuh, and P. Burlina “Automatic Segmentation of the Left-Ventricular
Cavity and Atrium in 3D Ultrasound Using Graph Cuts and the Radial
Symmetry Transform”, in Proc. IEEE Int. Symp. Biomed. Imaging, 2011.
R.
Mukherjee, C. Sprouse, T. Abraham, B. Hoffmann, E. McVeigh, D. Yuh, and
P. Burlina, “Myocardial Motion Computation In 4D Ultrasound”, in Proc.
IEEE Int. Symp. Biomed. Imaging, 2011.
D. Yuh, F. Contijoch, T. Ngo, D. Herzka, P.
Burlina, M. Brady, E. McVeigh, “3D-TEE Based Computational
Modeling of the Mitral Valve for Surgical Planning”, Proc. Conf. of
American Association for Thoracic Surgery, 2011.
P. Burlina, A. Pinhero, B.
Hoffmann, D. Yuh, T. Abraham, “Estimating Blood Flow Motion In The Left
Intraventricular Cavity”, in Proc. Int. Conf. Applied Bionics and
Biomechanics, 2010.
P. Burlina, C.
Sprouse, A. Jorstad, D. DeMenthon, F. Contijoch, T. Abraham, E. McVeigh, R.
Juang, and D. Yuh, “Individualized Cardiothoracic Surgical Planning using
Computer Aided 3D Modeling and Image Analysis”, in Proc. first International
AMA-IEEE Medical Technology Conference on Individualized healthcare, 2010.
P. Burlina, D.
Freund, B. Dupas, Shiri Zayit-Soudry, N. Bressler, “An Automated System for Prescreening
Individuals with Retinal Abnormalities”, in Proc. first International AMA-IEEE
Medical Technology Conference on Individualized healthcare, 2010.
P. Burlina, A.
Pinhero, D. Yuh, T. Abraham, “Estimating Blood Flow
Motion In The Left Intraventricular Cavity”, in Proc.Int. Conf. Applied Bionics and Biomechanics, 2010.
A. Banerjee,
R. Juang, J. Broadwater, and P. Burlina, “Sparse Feature Extraction for Support
Vector Data Description Applications”, in Proc. IEEE IGARSS, 2010.
H.V. Nguyen,,
A. Banerjee, P. Burlina, J. Broadwater, R. Chellappa, "Tracking via Object
Reflectance using a Hyperspectral Camera", CVPR Object Tracking and
Classification Beyond and in the Visible Spectrum workshop 2010.
A. Jorstad, P.
Burlina, I-J. Wang, D. Lucarelli, D. DeMenthon, R. Tron, A. Terzis, R. Vidal, “Estimating the Pose of an
Object in a Distributed Camera Network”, in JHU APL Tech Digest, Vol 8, No 3, 2010.
C. Sprouse, A.
Jorstad, D. DeMenthon, P. Burlina, F. Continjoch, T.
Ngo, D. Herzka, E. McVeigh, J. Stearns, K. Grogan, M. Brady, T. Abraham and D.
Yuh, “Computational Cardiac Modeling Based on Transesophageal Echocardiographic
Imaging”, in JHU APL Tech Digest, Vol 8, No 3, 2010.
D. Freund, P.
Burlina, and N. Bressler, “A Machine Learning Approach to the Detection of
Intermediate Stage of Age-related Macular Degeneration”, in Proc. Int. Conf.
American Association for Research in Vision and Ophthalmology, 2009.
D. Freund, P.
Burlina, and N. Bressler, “Automated Detection of Drusen in the Macula”, in
Proc. IEEE Int. Symp. Biomedical Img., 2009.
R. Juang and
P. Burlina, “Comparative Performance Evaluation of GM-PHD in Clutter”, in Proc.
IEEE Int. Conf. Fusion 2009.
D. Freund, P.
Burlina, A. Banerjee, and E. Justen, “Comparison of Kernel-Based Estimation
Methods”, Proc. SPIE Automatic Target Recognition Conference, 2009.
A. Jorstad, P.
Burlina, I-J. Wang, D. Lucarelli, D. DeMenthon, R. Tron, A. Terzis, R. Vidal, “Estimating the Pose of an
Object in a Distributed Camera Network”, APL S&T Research Symposium, 2009.
A. Jorstad, P.
Burlina, I.J. Wang, D. DeMenthon, D. Lucarelli, “Model-Based Pose Estimation by
Consensus”, Proc. Int. Conf. Intelligent Sensors, Sensor Networks and
Information Processing, 2008.
D. Freund, P.,
Burlina, and A. Banerjee, “Characterization of Spatial Ordering of Corneal
Fibers”, in Proc. IEEE Int. Symp. Biomed.
Img.,
2008.
R. Juang, P.
Burlina, and A. Banerjee, “Level Set Segmentation of Hyperspectral Images Using
Joint Spectral Edge and Signature Information”, in Proc.11th
International Conference on Information Fusion, 2008.
R. Meth, A.
Banerjee, P. Burlina, and T. Strat, “Rapid, High
Performance Hyperspectral Anomaly Detection via Global Support Vector Data
Description”, Proc. SPIE, 2008.
A. Banerjee,
P. Burlina, and R. Meth, “Fast Hyperspectral Anomaly Detection Via SVDD”, Proc.
IEEE Int. Conf. Image Processing, 2007.
A. Banerjee,
P. Burlina, and R. Meth, “Fast Hyperspectral anomaly detection using SVDD”,
Proc. IEEE ICIP, San Antonio, TX, 2007.
C. Sprouse, P.
Burlina, and R. Awadallah, “Polarization-Based High Resolution Radar Scatterer Classification”, Fourth IEEE (SAM-2006), Waltham,
Massachusetts 12-14 July 2006.
C. BenAbdelkader, P. Burlina, and L. Davis, “Single camera
multiplexing for multi-target tracking”, Proc. of IEE Int'l Conf. on Image
Processing and its Application (IPA '99), pp. 1140-1143, 1999.
A. N. Rajagopalan, P. Burlina, and R. Chellappa, “Higher order
statistical learning for vehicle detection in images”, Proceedings of the
Seventh International Conference on Computer Vision, volume 2, pages 1204-1209,
1999.
A. N. Rajagopalan, P. Burlina, and R. Chellappa, “Detection of
people in images”, Proc. International Joint Conference on Neural
Networks, Vol. 4, 10-16, Pages: 2747–2752, vol. 4, 1999.
P. Burlina,
“MPEG-Domain Sprite Generation”, Proc 10th International Multi Dimensional
Signal Processing, July 12-16, 1998, Alpbach,
Austria, 1998.
C. Shekhar, S.
Moisan, R. Vincent, P. Burlina, and R. Chellappa,
“Use of Knowledge-Based Control for Vision Systems”, Proceedings of the 11th
International Conference on Industrial and Engineering Applications of
Artificial Intelligence and Expert Systems: Tasks and Methods, pp. 427-436,
1998.
A. Banerjee,
P. Burlina, R. Chellappa, and R. Kapoor, “Frequency
dependence of ATD performance in foliage-penetrating SAR images”, Proc.
International Conference Image Processing, Vol. 1, pp5 78 – 582, 1998.
G.
D. Cetintemel and P. Burlina, “On-the-fly snake construction from
video”, Proc. IEEE International Conference on Image Processing, pp
638–642, vol. 3, 1998.P. Burlina and. D. Cetintemel, “On-the-Fly Road
Tracking from video”, Proc. DARPA IU Workshop, 1998.
R. Chellappa,
P. Burlina, L. Davis, and A. Rosenfeld, “SAR/EO Vehicular Activity
Analysis System Guided by Temporal and Contextual Information”, Proc.
DARPA IU Workshop, 1998.
A. Banerjee,
P. Burlina, and F. Alajaji, “Contagion-based Image Segmentation and Labeling”,
Proc. International Conference on Computer Vision, (Bombay, India), January
1998.
F. Alajaji, S.
Al-Semari, and P. Burlina, “An unequal error
protection trellis coding scheme for still image communication”, Proceedings of
1997 IEEE International Symposium on Information Theory, 1997.
A. Banerjee
and P. Burlina, “Segmentation based Detection of targets in SAR”, Proc SPIE,
1997.
P. Burlina, V.
Parameswaran, and R. Chellappa, “Sensitivity analysis
and learning strategies for context-based detection algorithms”, DARPA Image
Understanding Workshop, New Orleans, LA, May 1997.
R. Chellappa,
Q. Zheng, S. Kuttikkad, C. Shekhar, and P. Burlina, “Site
Model Construction for the Exploitation of E-O and SAR Images”, in RADIUS:
Image Understanding got Imagery Intelligence, (eds.), Oscar Firschein
and Thomas Strat, Morgan Kaufmann Publishers, San
Mateo, CA, pp. 185-208, 1997.
V. Parameswaran, P. Burlina, and R. Chellappa, “Performance
Analysis and Learning Approaches for Vehicle Detection and Counting in Aerial
Imagery”, Proc. ICASSP 1997.
R. Chellappa,
Q. Zheng, C. Shekhar, and P. Burlina, “Site Model Supported Targeting”, DARPA
Image Understanding Workshop, New Orleans, LA, May 1997.
M. Srinivasan,
R. Chellappa, and P. Burlina, “Adaptive Source-Channel Subband
Video Coding for Wireless Channels”, Proc. of the First IEEE Workshop on
Multimedia Signal Processing, Princeton, NJ, June 1997.
C. Morimoto,
P. Burlina, and R. Chellappa, “Video Coding Using Hybrid Motion Compensation”,
Proc. International Conference on Image Processing, (Santa Barbara, CA),
October 1997.
F. Alajaji, S.
Al-Semari, and P. Burlina, “An Unequal Error
Protection Scheme for Still Image Communication”, Proc. International Symposium
on Information Theory, (Ulm, Germany), June 1997.
C. Shekhar, P.
Burlina, and S. Moisan, “Design of self-tuning IU
systems”, DARPA Image Understanding Workshop, May 1997, vol. 1, pp. 529--536,
New Orleans, LA, 1997.
P. Burlina, C.
L. Lin, and R. Chellappa, “On a Spectral Attentional
Mechanism”, Proc. of Conference on Computer Vision and Pattern Recognition,
(San Francisco, CA), June 1996.
C. Morimoto,
P. Burlina, R. Chellappa, and Y. S. Yao, “Performance Analysis of Model-Based
Video Coding”, Proc. International Conference on Image Processing, (Lausanne),
September 1996.
F. Alajaji, P.
Burlina, and R. Chellappa, “Map decoding of gray-level images over binary
channels with memory”, Proc. International Conference on Image Processing
(Lausanne), September 1996.
X. Zhang, P.
Burlina, Q. Zheng and R. Chellappa, “Automatic Image to Site Model
Registration”, Proc. International Conference on Acoustics, Speech and Signal
Processing, (Atlanta, GA), June 1996.
R. Chellappa,
S. Kuttikad, R. Meth, P. Burlina, K. Eom and C. Shekhar, “Model Supported Exploitation for
Synthetic Aperture Radar Images”, Proc. of ARPA IU Workshop, (Palm Springs,
CA), 1996.
Y. Yao, P.
Burlina, and R. Chellappa, “Stabilization of Images Acquired by Unmanned Ground
Vehicles”, Proc. of ARPA IU Workshop, (Palm Springs, CA), 1996.
R. Chellappa,
X. Zhang, P. Burlina, Q. Zheng, C. L. Lin and A. Rosenfeld, “An Integrated
System for Site Model Supported Change Detection”, Proc. of ARPA IU Workshop,
(Palm Springs, CA), 1996.
R.
Chellappa, S. Kuttikkad, R. Meth, P. Burlina, and C. Shekhar,
“Model-supported exploitation of synthetic aperture radar images”,
Proc. SPIE, 1996.
P. Burlina, F.
Alajaji, and R. Chellappa, “Transmission of Two-tone Images over Noisy Channels
with Memory”, Proc. International Conference on Information Theory, (Vancouver,
BC), June 1995.
F. Alajaji and
P. Burlina, “Image Modeling and Restoration Through Contagion Urn Schemes”,
Proc. International Conference on Image Processing, (Crystal City, VA), June
1995.
R. Chellappa,
P. Burlina, S. Kuttikkad, C. L. Lin, and X. Zhang,
“Context-Based Exploitation of Remotely Sensed Imagery”, Proc. Context-Based
Vision Workshop, (Cambridge, MA), June 1995.
Y. S. Yao, P.
Burlina, R. Chellappa, and T. H. Wu, “Electronic Image Stabilization from
Integrated Visual Cues”, Proc. International Conference on Image
Processing" (Crystal City, VA), June 1995.
R. Chellappa,
Y. S. Yao, and P. Burlina, “Processing and Understanding of Image Sequences”,
Proc. Imaging Science and Technology Annual Conference, Washington, D.C., May
1995.
P. Burlina and
R. Chellappa, “Spatio-temporal moments and
generalized spectral analysis of divergent images for motion estimation”, Proc.
IEEE International Conference Image Processing, Vol. 1, pp. 328 – 332, 1994.
P. Burlina and
R. Chellappa, “Time-to-X: Analysis of Motion through Temporal Parameters”,
Proc. of Conference on Computer Vision and Pattern Recognition, (Seattle, WA),
June 1994.
P. Burlina and
R. Chellappa, “Spectral Separability of Divergent
Image Motion”, Proc. of the Conference on Information Science and Systems,
(Princeton University), 1994.
P. Burlina and
R. Chellappa, “Monocular Detection of Order, Collision and Other Kinematic
Events”, Proc. of the IEEE/IES Symposium on Intelligent Vehicles, (Paris,
France), October 1994.
P. Burlina and
R. Chellappa, “Spectral and Temporal Representations of Looming and Maneuvering
Information”, Proc. of ARPA IU Workshop, 1199-1207, (Monterey, CA), 1994.
PATENTS
US20020176604, US20040225730, US20040216084, WO2010/071898TEACHING
525.443 Real Time Computer Vision
E. McVeigh | F. Alajaji | R.
Chellappa | D. DeMenthon | N. Bressler | T. Abraham | B.
Hoffmann | A. Levchenko |
R. Vidal
| D. Yuh | C. Sprouse | B. Dupas
p h i l ( a t ) p m b u r l i n a (d o t) c o m
Copyright © P.Burlina 2009-2011