The document discusses the growing issue of large medical imaging data. It notes that imaging data sizes are rapidly increasing due to factors like higher resolution scans and more scans being performed. This is creating challenges for data storage, management, transfer and analysis. The document provides examples of typical data sizes for different types of scans and estimates for total imaging data storage needs. It also discusses the DICOM standard and its role in managing medical imaging data but notes challenges it faces with large data sizes and network transfers.
8. Positron Emission Tomography
• Functional imaging
• Radioactive bio-marker binds to
cancerous cell
PET-CT • Capture positron decay with a
scintillation detector
9. Basic Imaging Chain
The fancier the DAS,
The larger the data
Digital Acquisition System
Image Reconstruction
Add Patient Information
10. How BIG is that Image Data ?
Header Primarily “Text”
Few KB
Pixel
1024 x 1024
16 bits per pixel
2 MB
(tchah ! That’s not much)
11. They don’t come alone !
Header
Header
Header
Pixel Header
Pixel Header
Pixel Header
Pixel Header
Pixel
Pixel
Pixel
Images come in “Stacks”, a.k.a
Volumes
Series
Slices
12. How BIG is that Image Data ?
Full body PET CT Cardiac fMRI
Images / set 600 3000 20000
Size of 1 set 1.2 GB 6 GB 40 GB
No. of sets (typical) 4 6 8
Exam Size 9 GB 36 GB 300 GB
Sizes are approximations
13. How do we take that data home ?
…and who does it belong to ?
14. To store and share
digital data,
we need a
format
and a
protocol
15. And there was none until 1993 !
“And the whole earth
was of one language and
of one speech” ~ Genesis 11
17. DICOM Scope
Medical
Informatics
Patient
Bedside
...
Monitoring
Administrative
HIS/RIS Lab Data
... Diagnostic
Imaging
Scope of
DICOM
18. DICOM Standard
20 PS 3.1:
PS 3.2:
PS 3.3:
Introduction and Overview
Conformance
Information Object Definitions
parts PS 3.4:
PS 3.5:
Service Class Specifications
Data Structure and Encoding
PS 3.6: Data Dictionary
PS 3.7: Message Exchange
PS 3.8: Network Communication Support for Message Exchange
+
PS 3.9: Point-to-Point Communication Support for Message Exchange (Retired)
PS 3.10: Media Storage and File Format for Data Interchange
PS 3.11: Media Storage Application Profiles
PS 3.12: Storage Functions and Media Formats for Data Interchange
PS 3.13: Print Management Point-to-Point Communication Support (Retired)
PS 3.14: Grayscale Standard Display Function
PS 3.15: Security Profiles
161
PS 3.16: Content Mapping Resource
PS 3.17: Explanatory Information
PS 3.18: Web Access to DICOM Persistent Objects
supplements
PS 3.19: Application Hosting
PS 3.20: Transformation of DICOM to and from HL7 standards
19. DICOM Network:
where is it in the network stack?
Medical Imaging Application
OSI upper layer
Service boundary
DICOM Application Entity
OSI Association Control
DICOM Service Element (ACSE)
Upper Level
Protocol OSI Presentation Kernel
for TCP/IP OSI Session Kernel
TCP/IP OSI Transport OSI
stack TCP stack
OSI Network
IP LLC
Standard Network Physical Layer
(i.e. Ethernet, FDDI, ISDN, etc.)
21. How much disk space is needed for
cardiac screening of Bangalore ?
Back of the envelope calculations
22. Population of Bangalore : 7,000,000
Cardiac Screening : 50%
HDD / CT Cardiac : 20 GB
I need some space : 70 PB
23.
24. Some more size estimates
Christian Medical College Vellore 60 TB
0.5 million exams / yr
Clalit Healthcare Services, 250 TB
14-hospital network in Israel
4.5 million exams/yr
(annually)
Est. imaging data size in US – 2014 100 PB
Est. imaging data size globally – 2020 35 ZB
26. Lawmakers demand storage guarantee
Moms:
“25 years after the birth of the last child”
Mentally disabled:
“20 years after the last contact or 8 years
after the patient's death”
Children:
“Until the patient is 25”
Storage Commitment built into DICOM
27. Huge Capacity requirements
Huge CapEx
Unpredictable TCO
Hardware technology obso
Space
Data-center grade infra
28. Finding the needle in the haystack
Simple search is a soft problem
Index only the meta-data
DICOM loves SQL
29. SQL Tables
Relevant header info
Header
Header
Header
Pixel Header
Pixel Header
Pixel Header
Pixel Header
Pixel
Pixel
Pixel
Complete File
Flat Files
Power of SQL queries
Insertion is fast, Read is fast
Replicate tables
Memory-mapped IO
Better disaster recovery
30. Why move the data ?
• Offline storage [Store/Fetch]
• Reporting
• Teleradiology / Remote reporting
32. Why move the data ?
• Offline storage [Store/Fetch]
• Reporting
• Teleradiology / Remote reporting
33. Challenges of DICOM
• DICOM is based on TCP/IP
• Slow over large number of hops
• FileCatalyst, CISCO WAAS
• DICOM compression is not adequate
• Lossy, Loseless
• DICOM is not efficient on fault-tolerance
• Dated retry mechanism, transmit in
sets/series, not files
34. FOSS DICOM Tools and Images
Viewers
API
OsiriX for Mac
Language Toolkit Santesoft for Win
C/C++ GDCM, DCMTK Kradview for Linux
Java Pixel, dmc4che
Perl DICOM.pm
Ruby Ruby DICOM Public Datasets
Python pydicom
PHP Nanodicom ftp://medical.nema.org/medical/dicom/DataSets/
C# DICOM# http://www.barre.nom.fr/medical/samples/
Editor's Notes
Health is at the fundamental of human existence, literally. For time immemorial, mankind has been trying to get a look into what is inside the human body.
In 200 years, we have a come a long way. Information is now digital. And visual. You can really see inside the human body, not just hear it.