Biomedical Computing and Visualization

1
Problem Solving Environments for DDDAS
Chris Johnson Scientific Computing and Imaging
Institute University of Utah
2
SCI Institute
3
(No Transcript)
4
DOE ASCI Center C-SAFE
5
C-SAFE - End to End Simulation
6
(No Transcript)
7
Visualization Software BioImage
8
Time Dependent Visualization
Time-dependent BioImage volume rendering of a 4D
CT dataset. Interaction rate on a PC running
Linux is approximately 10 frames per second.
Data from collaborator George Chen (MGH).
9
DDDAS Challenges

• DDDAS projects offer unique challenges
• Complex software integration
• Multiple programming languages
• Inherently distributed
• Parallel algorithms
• Interdisciplinary
• Dynamic
• Need to change algorithms dynamically

10
Sensor Networks

• Sensor networks are distributed systems
• Typically have limited computational power
• Stream data to larger computational resources

11
Introduction to Components

• Good Fences make Good Neighbors
• A component architecture is all about building
  (and sometimes enforcing) the fences
• Popular in the software industry (Microsoft COM,
  CORBA, Enterprise Java Beans)
• Commercial component architectures not suitable
  for Scientific Computing (CCA Forum organized to
  address this point)
• Visual programming sometimes used to connect
  components together

12
Software SCIRun, BioPSE, PowerApps

• Computational Workbench
• Image Processing, Modeling, Simulation,
  Visualization Tools
• Layered Software System
• Dataflow/Modules
• PowerApps
• Interactive Problem Set up and Computing
• What if..? Problem Exploration
• Flexible, Extensible Environment
• Open Source Software (MIT Public License)
• Module Development Tool and Bridging
• PowerApps (BioFEM, BioTensor, BioImage)
• Irix, Linux, Mac OSX, Windows

13
SCIRun

• Modules - functional elements
• Computation
• Visualization
• Modeling
• Imaging
• Data flows through ports
• Fields - the main data model
• Nets - are the application

14
SCIRun - Networks
Point Cloud
Field
Scan line
Curve
TriSurf
QuadSurf
LatVol
Image
Matrix
Geometry
TetVol
HexVol
15
SCIRun/BioPSE Extensibility

• Leverage existing utilities/programs
• Extensibility through bridges

GENESIS
Matlab
NetSolve
SQL
lib
socket
BioPSE
PETSc
lib
ITK
lib
16
ITK An Interactive Segmentation Tool
www.itk.org
17
(No Transcript)
18
Cardiac Electrical Wave Propagation
Structural and chemical information
Cellular model of electrical conductivity/ action
potential propagation
Bidomain model of an ischemic heart
19
Software Tool Integration
Computational Cardiology Software Integration
Example
Matlab for cellular modeling and some numerics
Matlab tools
CardioWave is specially designed to run on a
cluster to compute bidomain simulations
CardioWave Simulation tool
SCIRun visualization
SCIRun for visualization, imaging, integration
20
SCIRun
gt Data flow paradigm gt Integrating
different programs together gt Advanced
visualization
21
(No Transcript)
22
(No Transcript)
23
(No Transcript)
24
(No Transcript)
25
DDDAS Example

• Sensor processing performed in Matlab
• Visualization and analysis performed in SCIRun
• Other components being added

26
Software Engineering
Software Engineering is Often Overlooked! Software
Engineering is Hard! Software Engineering is
Important!
27
Information Big Bang
28
New Visualization Techniques
29
VisTrails Visualization Meets Databases

• Create and compare large numbers of
  visualizations
• Separation between dataflow definition and
  instances
• Optimized execution
• Multi-view visualization
• Comprehensive support for exploratory process
• Vistrail Evolving dataflow
• Keep data alive
• Unobtrusive tracking of visualization provenance
• Reproducible results
• Everything in XML
• Open standards allow easy interoperability
• Sharing and collaboration
• Visualization metadata can be queried and mined

30
VisTrails Archicture
31
VisTrails History Management
A single VisTrail is a complete collection of
dataflows!
32
VisTrails Multi-View Functionality
33
NSF Report on Simulation Based Engineering Science
http//www.ices.utexas.edu/events/SBES_Final_Repor
t.pdf
34
PITAC Report www.nitrd.gov
35
NIH/NSF Visualization Research Challenges Report
http//tab.computer.org/vgtc
36
(No Transcript)
37
(No Transcript)
38
(No Transcript)
39
Acknowledgements

• NIH NCRR and BISTI
• DOE ASCI and SciDAC
• NSF
• SGI Visual Supercomputing Center
• Utah Centers of Excellence
• Visual Influence

40
More Information
www.sci.utah.edu
41
More Information

• www.sci.utah.edu
• crj_at_sci.utah.edu