; Pan GridCAD
Documents
Resources
Learning Center
Upload
Plans & pricing Sign in
Sign Out
Your Federal Quarterly Tax Payments are due April 15th Get Help Now >>

Pan GridCAD

VIEWS: 3 PAGES: 20

  • pg 1
									  gridIMAGE Microscopy: A caBIG
Based System for Image Processing
    and Quantitative Analysis

     Tony Pan, Ashish Sharma, Metin Gurcan
      Kun Huang, Gustavo Leone, Joel Saltz

     The Ohio State University Medical Center, Columbus OH




           For more information, please contact Tony Pan (tpan@bmi.osu.edu)
      Dept. of Biomedical Informatics, The Ohio State University http://bmi.osu.edu
                 Agenda

•   Motivation
•   caGrid overview
•   gridIMAGE Radiology
•   gridIMAGE Microscopy
•   Future Directions
  Digitized Microscopy: Virtual Slide
          Cooperative Studies
• CALGB, Children’s
  Oncology Group
  Cooperative Studies
• Roughly 30 slides/day –
  30 GB/day compressed,
  300GB/day
  uncompressed
• Remote review of slides
• Tissue bank QA/QC
• Computer assisted tumor
  grading
        EXAMPLE: Large Scale Imaging Pipeline
      Con-focal Microscopy (joint work with NCMIR)

                             correctional tasks
                Image file
                                normalization      stitching         warping


                                                                   declustering
                  target task      preprocessing tasks

                                     prefix sum
                     querying                       tessellation   thresholding
                                     generation



• Problem definition: how many pixels of a certain color intensity
  exist within a rectilinear region of interest?
• Implementation: the prefix sum solves the query without
  scanning every pixel within the region of interest
                       What is Grid?
• A lot of different things to a lot of different people
• Evolution of distributed computing to support sciences and
  engineering
• Some common themes prevail:
   – Sharing of resources (computational, storage, data, etc)
   – Secure Access (global authentication, local authorization, policies, trust,
     etc)
   – Open Standards
   – Virtualization
• ―The real and specific problem that underlies the Grid concept
  is coordinated resource sharing and problem solving in
  dynamic, multi-institutional virtual organizations.‖
   – I. Foster, C. Kesselman, S. Tuecke. International J. Supercomputer
     Applications, 15(3), 2001.

• A good general overview can be found here:
  http://gridcafe.web.cern.ch/gridcafe/
              What is caGrid?

• Development project of NCI caBIG Architecture
  Workspace, aimed at helping define and
  implement Gold Compliance
• No requirements on implementation technology
  will be necessary for Gold compliance
  – Specifications will be created defining requirements
    for interoperability
  – caGrid provides core infrastructure, and tooling to
    provide ―a way‖ to achieve Gold compliance
• Gold compliance creates the G in caBIG
  – Gold => Grid => connecting Silver Systems
           Benefits and Motivation

• Facilitate research and clinical decision support with large number of
  datasets and multiple analysis algorithms.
    – Parameter studies, clinical and preclinical trials, etc
• Enable better algorithm development and validation through the use of
  many distributed, shared image datasets
• Support remote algorithm execution – reduce data transfer and avoid the
  need to transmit PHI
• Reduce overall processing time and algorithm development cycle through
  remote compute resource recruitment and CAD compute farms
• Scalable and open source — caGrid 1.0 based


               Data and Algorithm Sharing over the Internet
 gridIMAGE Radiology
Expose algorithms, human markup and
   image data as caGrid Services
                   Image Data Service
• Expose data in PACS servers as caGrid Data Service
• Open source DICOM server — PixelMed
• XML based data transfer (NCIA-like schema)

     3 Participating Data Services




                    Columbus

                                     caBIG
  Los Angeles
                CAD Application Service
 •   caGRID middleware to wrap CAD applications with grid services
 •   Interact with Data Services to retrieve images
 •   Invoke algorithm with required inputs
 •   Transform and report results to results data service
     2 Participating Analytic Services




                        Columbus

                                        caBIG




caGrid Introduce Hides complexity of                               caGrid Dorian Used to
plugging an algorithm into the grid                                provide authentication service
           CAD algorithms provided by iCAD Inc. Prototypes for investigational use only; not commercially available
           Human Markup Services
• Query a work-order queue to detect any new markup requests
• Interact with Data Services to retrieve images
• Capture markups and save to results data service




       2 Human Markup Services

                            Baltimore
                 Columbus
            User Interface

Available
data                                        DICOM
services                                    image viewer




Queried
results




                    Click to browse images, submit
                    CAD analysis, and view results
                 Technologies

• caBIG caGrid 1.0 beta
  – Globus Toolkit 4.0.1 compliant
  – Introduce toolkit for service creation and deployment
  – Dorian security management for user and service
    authentication and authorization
  – CQL based query and retrieve for data services
• External applications and algorithms
  – Matlab
  – Lung Nodule CAD
  – etc
           gridIMAGE Microscopy

• A prototype implementation to
  demonstrate applicability of
  gridIMAGE Radiology
  architecture for microscopy
  image analysis

• Liver macrophage
  quantification
   – IHC staining
   – Single field of view capture in
     JPEG format
   – Matlab algorithm for
     segmentation and
     quantification
            gridIMAGE Microscopy Architecture

•   The Image Data Service holds
    microscopy images
                                                 Image
     –   caGrid Image retrieval via SOAP and    Storage
         Java object serialization
     –   Data modeled using XML schema
•   Application Service
     –   Interfaces with Matlab server to        Matlab
         execute algorithms                     Algorithm

     –   retrieves images directly from Image
         Data Service
•   Result handling
     –   images are submitted back to the
         Image Data Service
     –   Return quantitative results to user
         interface
•   Current user interface support
     –   Command line based invocation
         currently
     –   GUI based image review and analysis
         invocation is next
Some Sample Results
           Benefits and Motivation

• Facilitate research and clinical decision support with large number of
  subjects and multiple analysis algorithms.
    – Parameter studies, clinical and preclinical trials, etc
• Enable better algorithm development and validation through the use of
  many distributed, shared image datasets
• Support remote algorithm execution – reduce data transfer and avoid the
  need to transmit PHI
• Reduce overall processing time and algorithm development cycle through
  remote compute resource recruitment and CAD compute farms
• Scalable and open source — caGrid 1.0 based


               Data and Algorithm Sharing over the Internet
                           Future Direction
• Usability
     GUI support for microscopy image review
     Whole slide image support

• Advanced algorithms
     More real-world algorithms for real applications
     Distributed algorithms

• Location independence
     Move algorithms to data
     Move both data and algorithms to compute servers
     Currently supported – ongoing collaborations to deploy these capabilities

• Security and Privacy
     Encryption, authorization, and Just-In-Time anonymization for the image data services

• Scaling and Deployment
     High performance image transfer mechanisms
     Greater number and variety of image analysis algorithms
              Acknowledgements


  This project was funded by NIH BISTI Center for Grid
Enabled Medical Imaging, NCI, NSF, and the State of Ohio
            Board of Regents BRTT program




             For more information, please contact Tony Pan (tpan@bmi.osu.edu)
        Dept. of Biomedical Informatics, The Ohio State University http://bmi.osu.edu

								
To top