Transcript DataGRID Assurance Quality

ACI GRID 2002
MEDIGRID: high performance medical
image processing on a computational grid
(accepted 3 years project)
CREATIS (INSA Lyon)
J. Montagnat
LISI (INSA Lyon)
L. Brunie
ERIC (University Lyon 2)
S. Miguet
MEDIGRID project, DataGrid FR meeting, April 18, 2002, Johan Montagnat, WP10
Purposes
 Medical



image storage and processing on the grid
Structured data
Strong semantic (distributed medical files)
Sensitive (security issues)
 Process
complex algorithms with large computing power and memory
requirements

parallel processing
 Handle


very large data sets
distributed storage
massive distributed processing
 Context:
datagrid testbed.
MEDIGRID project, DataGrid FR meeting, April 18, 2002, Johan Montagnat, WP10
Scientific challenges
 Shared

data hierarchy, dynamic indices, optimization, caching
 Hybrid

and distributed data management
content-based and metadata queries
content-based queries, pre-processing, distributed queries
 Complex

modeling of anatomical structures
anatomical and functional models, parallelisation
 Simulation

of MRIs
MRI modeling, artifacts modeling, parallel simulation
MEDIGRID project, DataGrid FR meeting, April 18, 2002, Johan Montagnat, WP10
Data and users
 Medical



Data
Images and metadata
Nominative (critical) and non-nominative (private) data
DICOM3 for medical images
 Users



Patient: has free access to its medical data.
Physician: has complete read access to its patients data. Few persons have
read/write access.
Researchers: may obtain read access to anonymous medical data for research
purposes. Nominative data should be blanked before transmission to these
users.
MEDIGRID project, DataGrid FR meeting, April 18, 2002, Johan Montagnat, WP10
Data management requirements
 Medical




data requirements
Very large databases (Tb/year/hospital) and very long term storage (20 years
for all images, up to 70 years in some cases)
Large amount of metadata
Access right on a medical department basis
Log data processing
 Security





related requirements
Data access security (Read-only access for most users, access right on a
medical department basis, no read access to private data by any third party
including system administrators)
Explicit control of sites where private data may be replicated
Nominative metadata should be stored in secure databases
Images should not be replicated outside hospital without blanking headers
Images content should be encrypted
MEDIGRID project, DataGrid FR meeting, April 18, 2002, Johan Montagnat, WP10
Data infrastructure testbed
4
image sources:




Acquired from Lyon cardiological hospital
Bone structure database from ESRF Grenoble
Mammographies from the DDSM
Simulated images
INSA
Bone
Simulated
MRI
Cardiological
hospital
Heart
MRI
DLT
Breast
images
Local
cluster
University
Lyon 2
EDG
MEDIGRID project, DataGrid FR meeting, April 18, 2002, Johan Montagnat, WP10
Foreseen medical data infrastructure
 Split
nominative and anonymous data to allow data replication on
unsecured sites.
Untrusted data servers
with blanked images
hospital
jE
SE-DICOM
interface
SE
European
DataGrid
Authentication
and
Authorization
Header
blanking
DICOM
Server
Metadata
interface
SE
Confidential
data
Non-confidential
data
MEDIGRID project, DataGrid FR meeting, April 18, 2002, Johan Montagnat, WP10
Interaction with DataGrid middleware
WAN
qmsg/shm
SE Interface
jE Cache
hdS
security (GSI)
spitfire
hdS
jE Dicon Server
LAN
jE Spitfire Int
jE Dicon Client
OK
phase 1
phase 2
jE User Interface
jE/SE Interface
ATF April 8, 2002, H. Duque, WP10
Applications
 Image


database importation
Pre-processing of images for database indexing
Data distribution
 Hybrid


Triggering parallel processing, distributed queries, DICOM server interface
pipeline processing
 MRI

simulation
parallel simulation kernel
 Heart

content-based and metadata queries
segmentation from MRI sequences
heart modeling, parallel job execution
MEDIGRID project, DataGrid FR meeting, April 18, 2002, Johan Montagnat, WP10
Application example: Metadata query and
image processing
MEDIGRID project, DataGrid FR meeting, April 18, 2002, Johan Montagnat, WP10
Schedule
 year
1

setting up databases

statistical parameters computation

image indexing

spitfire-based security

MRI simulator

grid interface to DICOM servers
 year
2

interface with datagrid / remote processing

distributed data management

spitfire extensions

content-based and metadata queries

pipeline processing
MEDIGRID project, DataGrid FR meeting, April 18, 2002, Johan Montagnat, WP10
Schedule
 year
3

distributed data management

biomedical modeling

mammography studies

cardiac studies

first medical assessment
MEDIGRID project, DataGrid FR meeting, April 18, 2002, Johan Montagnat, WP10