SLIDE 1
Industry Trial Data: Mountains To Mine For AI Gold
Gregory Goldmacher, MD, PhD, MBA Executive Director, Head of Clinical Imaging Merck Research Laboratories
#FDAPDSsymp | #AIinHealth
Industry Trial Data: Mountains To Mine For AI Gold Gregory - - PowerPoint PPT Presentation
Industry Trial Data: Mountains To Mine For AI Gold Gregory - - PowerPoint PPT Presentation
Industry Trial Data: Mountains To Mine For AI Gold Gregory Goldmacher, MD, PhD, MBA Executive Director, Head of Clinical Imaging Merck Research Laboratories #FDAPDSsymp | #AIinHealth Outline Ecosystem and flow Data available AI
SLIDE 2
SLIDE 3
Outline
- Ecosystem and flow
- Data available
- AI opportunities
- Challenges
SLIDE 4
The Players
- Pharma companies (“sponsors”)
- Hospitals (“sites”)
- Independent review facilities (“iCROs”)
- Regulators
SLIDE 5
Site Sponsor iCRO
Protocol Manual Scans
Scan Scan Scan Scan BICR Responses Responses Responses Endpoints Stats Magic
Regulators
CSR
Data Flow
SLIDE 6
Analysis: Response Criteria
Baseline
Treat
Visit 1 Visit 2
Endpoints
Date of progression PFS Best response ORR etc… Lesion changes Visit 1 response Lesions Quantitative Qualitative
Treat
Lesion changes Visit 2 response
SLIDE 7
Read Process
- Quantitative
- Choose tumors to measure (baseline)
- Outline each on one slice (largest)
- Qualitative
- Judgment about “non-target” tumors
- Whether/when new tumors have appeared
- Synthesis
- Math and logic with human override
- “2+1” adjudication
SLIDE 8
Data Stored
- Images
- Scans
- Single slice outlines: every tumor on every scan
- Assessments
- Tumor locations and categories
- Measurements and qualitative judgments
- Calculated responses for every scan
SLIDE 9
Use In Training AI
- Human judgment
- Finding tumors
- Choosing what to measure
- Segmentation
- Response
- Non-imaging data
- Tissue/molecular
- Survival
SLIDE 10
Example 1 - Learning From Humans
- Manual segmentation is laborious
- 10,000 samples of single slice tumor outlines
- Trained CNN to segment tumors
SLIDE 11
Example 2 – Training on other data
- Gene expression profiling – inflammation signature
SLIDE 12
Obstacles To Sharing
- Technical
- iCRO holds the images, pharma owns them
- Older non-compatible data formats
- Sponsor concerns
- Re-analysis of efficacy
- New safety questions
- Imposed restriction of eligibility
SLIDE 13
Collaborate To Overcome Challenges
- Design smart datasets
- Trusted third parties
- Standard contracts / agreements
- Can regulators help “de-risk”?
SLIDE 14
Thank you!
#FDAPDSsymp | #AIinHealth