Web 2.0 and Beyond: Leveraging Web Technologies as Middleware in Healthcare Research

High Performance Compute Clusters: Data, Apps, Results, Sharing and Collaboration

 

Rudolph Pienaar, PhD
Technical Director
Fetal-Neonatal Neuroimaging and Developmental Science Center

Staff Scientist                                                      Instructor in Radiology
Boston Children's Hospital                         Harvard Medical School
 

Part I - Informatiomics in healthcare

• Forces acting on information
• Patterns of information flow
• Typical "apps"
• Rise of the walled information silos
• The coming information tsunami

Part II - Informatiomics and ChRIS

• Unmet need
• Core components
• Demo

External

• Regulatory
• Data ownership
  • Owned by individual, not institution

Internal

• Internal Review Board
• Locality
  • Department specific access
  • Idiosyncratic, isolated technolgoies 

Compounding Complexity

• Data is distributed in multiple places
  • EMRs
  • Modality Databases
  • Department databases
  • Different formats
• Data is structured and unstructured
  • "Voice" recordings
  • "Free form text"

Compounding Complexity

• Inconsistent/variable definitions
  • Quantitative interpretations
  • Qualitative measurement variability
    • T(R) on Siemens might be different to GE
• Data itself is complex
  • genomics
  • radiomics (images)
  • combinatorial explosion

Compounding Complexity

• Locked in isolation due to regulatory forces
  • Data movement out of institutions is complex
  • "Compute" movement into institutions is complex.

Act	Applies to
HITECH	Health Information Technology for Economic and Clinical Health Act, sets meaningful use of interoperable EHR adoption in the health care system as a critical national goal and incentivized EHR adoption.
Sarbanes-Oxley	Privacy and integrity of financial data in publicly traded corporations.
HIPAA	Confidentiality, integrity, and availability of health care information.
PCI	Confidentiality of credit card information stored and used by merchants.
GLBA	Confidentiality and integrity of personal financial information stored by financial institutions.
SB 1386	Confidentiality of customers' personal information stored by any organization that does business in the state of California.
BASEL II	Confidentiality and integrity of personal financial information stored by financial institutions. Availability of financial systems. Integrity of financial information as it is transmitted. Authentication and integrity of financial transactions.

Scope	Applies to
IRB	Internal committee formally designated to approve, monitor, and review biomedical and behavioral research involving humans.
Technology	Extremely static environment, typically on clinical side, with slow updates/changes to computational resources, especially as related to web-based technologies and mobile
Department	Supplementary technology is often department focused and department centric.

Current

• Unidirectional
• Isolated
• Non-integrative
• Static

Future

• Multi-directional
• Collaborative
• Integrative
• Dynamic

Current App Philosophy

• Old-style "web" apps subsumed to look like desktop apps.
  • Reflect an "app" centric model, not "data" centric.
• Not really "web" apps:
  • Typically limited to obsolete versions of Internet Explorer
  • No support for mobile or other platforms.
• No cross-app integration on data level.

• Typically no provision for "automated" or scripted interaction using modern techniques
  • REST API

FHIR: Fast Healthcare Interoperability Resources

• Defines an API for exchanging EHR
• HL7 compliant
• Uses modern web based APIs (HTTP/REST)
• JSON or XML for data representation
• Provides an alternative to document centric approaches by directly exposing discrete data elements as services

Square kilometer array

• Radio telescope being built in Australia and South Africa
• Total collecting area of one square kilometer
• 50 times more sensitive than any other radio instrument
• Data transport from dishes will produce 10 times the 2010 entire global internet traffic

SKA (daily estimated data generation)

• 1 exabyte
• 1,000 petabytes
• 1 million terabytes
• 1 billion gigabytes                                                              

Healthcare vs Financial

• Financial
  • Number of accounts: 10,000 to 300 million
  • Storage per Account: GB
  • Total Storage: 10s of TB to 300 PB
• Healthcare
  • Number of patients: 10,000 to 300 million
  • Storage per Patient: GB today, TB tomorrow
  • Total Storage: 20 PB to 600 EB

Healthcare

Healthcare

• 1.65M tumors measured in various modalities
• > 4 exabytes of data in various localities
• 1 petabyte of data (0.001EB) to transfer 
• On current Internet, would take about 21 days to transfer this data around.

ChRIS is:

• [Boston's] " Children's Research Informatics System"
• a web-based research/clinical workflow manager
• and more...

• collect and manage medical data with seamless integration with hospital systems
• process data
• visualize and interact with data
• share data
• real time collaboration

• Image analysis requires advanced scientific computing expertise

• unix commands
• disparate data and file systems
• disjointed analytical tools and modules

• lack of pipeline tools that combine processing modules, work flow integration, and a modern user interface
• high barrier to entry by non-technical users

• white matter tractography for surgical planning
• automatic brain surface reconstrution
• automatic brain volumetric segmentation
• automatic tumor detection and measurement

• specialized tools with high degree of domain knowledge
• UNIX (Linux or Mac) command line
• intimate knowledge of clusters and scheduling

• more familiar interface metaphor
  • web 2.0 "social" interaction
  • posting/sharing
• moves the burden of maintenance away from end user
• allows for mobile

• allows for advanced view rendering
• allows for sharing of images
• allows for immersive real time collaboration
• allows for post-processing of images

• collect
• manage
• analyze
• view
• collaborate

• seamlessly connect to standard PACS with Query/Retrieve
• direct access to server filesystem
• drag-n-drop from local client into server
• ssh/scp underpinning for secure network transfers

• all "data is stored in a time-based feed
• feeds are basically file-browsers
• feeds can be annotated with free form notes
• shared feeds can support comments between parties

• a "plugin" mechanism offers the ability to accept feed data and process data
• ChRIS is usually configured to communicate transparently with an HPC
• plugins are mostly non-interactive analysis pipelines
• interactivity is available as a pre-processing step

• image data in feed can be viewed directly in simple drop down window
• image data can be viewed in more sophisticated fashion with specialized viewing modes
• advanced sharing using 3rd party backends (Google Real Time API / Dropbox API)

• image data can be shared between collaborators
• Google RealTime API
• a realtime JSON object is shared and updates each viewer independently
• (akin to multi-player games)
• shared screen components (like cursor)

• ChRIS instance

• ChRIS instance

• https://fnndsc.github.io/mi2b2

• Jorge "To Be" Bernal, PhD
• Nicolas "VJ" Rannou, MSc
• Rudolph "Cat Herder" Pienaar, PhD

• Daniel "X Slicer" Haehn, PhD
• Daniel "Game On" Ginsburg, MSc

• Ellen "Rad" Grant, MD

• https://github.com/FNNDSC/chrisreloaded

• https://github.com/FNNDSC/ChRIS_API

• https://github.com/FNNDSC/viewerjs

• https://github.com/FNNDSC/mi2b2

• https://github.com/FNNDSC/vjs

• https://github.com/slicedrop

• https://goxtk.com