©2012, Cognizant
Examples of SOA and Cloud in Life Sciences
Sandeep Bhat
Vijay Srinivasan
Agenda
Objectives
• Industry Observations
• Four Technology Focus Areas for Success
• Canonical Illustrations
• Real World Cases : Four Examples
• Many opportunities for SOA and Cloud in Life Sciences
• SOA & Cloud Architectures promote newer business models
• Establish and Propagate reusable business services
• Create a Smarter Ecosystem
Consumerism
Patient Centricity, Changing Demographics, Personalized
Medicine
Build vs. Buy
Disintegration of the Value Chain, M&A, In-licensing
Cost of Innovation
R&D Investment, Value Differentiation
Just in Time
Care Delivery, Manufacturing
Pricing
Patent Expiration, Healthcare Reform
Commercial Operations
Newer Marketing and Sales Models
Digitally Enabled Silos
Biopharma (Sponsors)
Medical Technology /
HIT
Researchers / Prescribers / Physicians
Patients / Subjects / Caregivers
Regulators / Government
Payors / Insurance
Industry Observations
Game-Changing Trends and Massive Information Area Silos
Smart Federated
Cloud Services
Service Oriented
Composable Services
Security
Governance
Consumerism
Build vs. Buy
Cost of Innovation
Just in Time
Pricing
Commercial Operations
Four Focus Areas Central to SOA Success
• Enable newer business models
• High performance services (HPC, Big Data)
• Collaborative “Clouds “ - Harmonize Life Sciences value chains
Federated Cloud Architecture
• Reusable and composite services (Personalized patient care, clinical study service, Analytics) verticalized to all facets of the value chain
• Horizontal composite services – Regulatory Compliance services, secured EMR, federated IAM services . . .
Composable Services
• Federated IAM architecture across collaborative Life Sciences “Clouds”
• Patient EMR – Message Integrity, Message Confidentiality (SAML/WS-Security..)
Security
• Governance, Risk, and Compliance : for Audit and Control: 21 CFR Part 11 prescribes to both eSignature and Audit Trails, as when and necessary , functionally
• Corporate Integrity and Assurance Functions mandate that processes be institutionalized and Adhered.
Governance
The Four Focus Areas
Indicative Reference Architecture: On-Premise View
Note: This can be and will be technology-oriented
Line of Business
Channels
Presentation
Business Processes
Composite Services
Entity Services
Analysis
Stores 2 1
R D Mf Mk S
4 3 5
WEB MOBILITY OPERATIONS/HELP DESK
WEB NATIVE
SOCIAL
SOCIAL
OPERATIONAL REGULATORY STRATEGIC
TRANSFORMATIO
NAL
REUSABLE FEDERATED
ONTOLOGIES CANONICAL
INFORMATION MODEL
DISTRIBUTED
DATA STORES
STATISTICAL
MODELING MASHUPS WORKFLOW
Indicative Reference Architecture: Cloud Communities
Clinical
Real World Evidence
Tests and Biomarkers
Medical Records
Simulations Safety
Real World Evidence
(OMOP)
Signals
Sentinel
Simulations
Note: This will most likely be business function-oriented with a high compliance need
Regulatory
Risk Management
Multinational
Legal
Multiple Agency
Corporate Compliance
and Assurance
©2012, Cognizant
Cases
Two Case Studies
Two PoVs
Example 1: Safety Assessment Portal
Challenge
Challenges
• Inconsistent user experience
• Multiple interfaces
• Decentralized security
• Scattered information
• No self service functionality
• Lack of dashboard metrics
Global documentation system
Documentum
Legacy documentation system
Documentum
Clinical trials system
Oracle DB
Registry System
Oracle DB
Aggregate reporting system
BO
Clinical reports submission system
Oracle DB
Product labeling system,
Documentum
Regulatory reporting system
BO
Example 1: Safety Assessment Portal
Action
Provide a Product Centric view, through a portal, for safety Documents.
Provide ability for users to run parameterized reports and to view dashboards
Example 1: Safety Assessment Portal
Result
Minimize duplication, enhance
efficiencies
Enables users to view safety
information of products through
unified interface
Supports common analysis and
messaging between
documents
Product centric view of Safety documents
Example 1: Safety Analytics Portal
Result (screenshots)
Homepage Product Information Page
Report Generator NDA Dashboard
Example 2: Clinical Data Repository
Challenge
Context
FDA’s JANUS repository is a common, standards-based electronic infrastructure
that supports submission, validation, data warehousing, access and analysis of
clinical and non-clinical study data.
SDTM is the standard for the submission of data to JANUS. The goal of this effort
was to achieve a set of project objectives:
High Level Objective
1. Harmonize many different clinical information silos built using different &
evolving clinical standards to SDTM
2. Create ability to compare and aggregate data across studies and phases
3. Ensure consistent application of standards (SDTM) and processes to
improve productivity across the Clinical Data Life Cycle
4. Enable better collaboration with vendors, partners, and regulatory agencies
5. Enable next generation uses; e.g. translation medicine, study design inputs
Example 2: Clinical Data Repository
Action
1. Composable Architecture: A
Unified Information Model, pulls
information from multiple
sources
2. Governance and Compliance: 21
CFR Part 11, Version history
3. Future-Proof for Federated
Clouds: (e.g.: Genomics)
4. Security
1 Metadata Load (ODM)
2 Study level transfer of Metadata
3 Patient data load (SDTM)
4 Populate Metadata History
5 Transfer JANUS to JANUS History
6 JANUS History to SDTM repository
7 SDTM views from SDTM repository
8 SDTM + views (includes supp qualifiers)
9 Error handling service
10 Process logging service
11 Security access layer
12 UI to manage metadata, Data Set Generation and
Safety authoring process, Error remediation
13 Encoding service
14 Vocabulary update service
15 Derivation service
16 Service to generate datasets. Freeze service to
generate frozen files
Example 2: Clinical Data Repository
Result
Example 3: Investigator Research Centre Portal
Challenge
The TransCelerate Initiative
• Shared user interface for an investigator site portal
• Mutual recognition of Good Clinical Practice site training and qualification
• Risk-based monitoring approach and standards
• Clinical data standards for efficacy
• A system for ensuring safe and timely supply of comparator drugs
Making a Portal
• Two primary communities (Sponsor and Site)
• A future community (Subjects/Patients)
• Enable the tactical and strategic intents for research
• Standardized business functions (composable)
• On going and Active collaboration (just-in-time)
Two groups intents
Example 3: Investigator Research Centre Portal
Tactical
• Study Planning
• Contract Performance
• Enrollment
• Risk Management
Strategic
• Simulation
• Virtualization
• Labs (Imaging, ECG)
• Clinical Experience
Key Architecture Considerations
Reusable composite services for core clinical investigational services – Adopt SOA service
composition model to reuse functionality across sponsor, site
Composite UI Application Model – Aggregate and isolate specific user requirements catering to
sponsor, site
Reusable composite entity services – Abstract aggregation and assembling of relevant
information for multiple data sources – on premise legacy clinical apps/CDRs, ONTOLOGIES,
harvest evidence/clinical experiences from life sciences community Clouds
Action: Indicative Reference Architecture
Example 3: Investigator Research Centre Portal
Example 3: Investigator Research Centre Portal
Result – Future
Making the Business Smarter
• Greatly improved sponsor efficiencies
• Unification with site and true collaboration
• A move toward the patient “layer”
Application of SOA principles
• Service Composition through careful adoption of MEPs between service
composite and among composition members
• Service autonomy expressed thru careful adoption of MEPs and loosely
typed operations
• Services allow for both REST & SOAP messaging
• Vertical and Horizontal service compositions configurable– WS-* Primitives
• Shared Portal Services – Composite User Interface Applications
Objectives
• Many opportunities for SOA and Cloud in Life Sciences
• SOA & Cloud Architectures promote newer business models
• Establish and Propagate reusable business services
• Create a Smarter Ecosystem
External Associations
In house Associations
Data Exchange Services
Core Data
EDI Java Web Services ETL Archival Media FTP
Randomization Supply Subject
Identification
Master Data
Management
(MDM) Investigational
Product Demand Label
Modular Functions and Direct Applications
Sourcing Manufacture Distribution Real World
Tracking
Packaging,
Labeling Forecasting
Reconciliation
and Return
Which
sites,
Vendors
Controls,
On
Demand
Batch,
Lot, Time
Point,
Visit
Shelf Life
Reissue
Tracking
Shipment
Supplies,
blister or
vial
Reports
Destructi
on,
Notificatio
n
Interactive
Randomization
(IVRS, IWRS)
Clinical Trial
Management
(CTMS)
Commercial
Manufacturing
(MRP)
Facility / Depot
Investigational Product Inventory Management User Interface
Real world outlet
- Internet portal
SSH or DMZ
enveloped
Intranet or VPN
accessed portal
User Groups
Regulatory
Affairs
Quality
Assurance
Clinical
Inventory
Management
Manufacturing Clinical Ops
Study Managers,
CRAs, Medical
Advisors
Partners
(CSOs, CROs)
CMOs
Depots
Country Units,
International RA
Example 4: Clinical Supply Chain
Conceptual View