Building Privacy-First Healthcare AI Platforms
summary
Claims a working stack: a permissioned ledger storing only hashes for tamper‑evident consent, PBFT‑style consensus, and smart contracts that gate differential privacy budgets. Implements federated learning with secure aggregation, edge adapters that emit cryptographic receipts, attested training binaries, and realtime monitoring that quarantines poisoned updates. Roadmap uses four phased rollouts from low‑risk pilots to high‑stakes PHI, with automated policy failures and auditable lineage so compliance evidence exists before audits.
AI-written summary · about
chapters (11)
- 0:00Introduction and Objectives
- 1:16The Healthcare AI Privacy Paradox
- 2:20Distributed Nature of Healthcare Data
- 3:40Blockchain as a Trust Foundation
- 5:03Federated Learning Explained
- 9:03Privacy Engineering Integration
- 11:13DevOps Integration and Compliance
- 13:20Multi-Institutional Scaling Challenges
- 14:33Platform Monitoring and Observability
- 15:44Implementation Roadmap
- 17:34Conclusion and Final Thoughts
related talks
24mAI, Health & Confidentiality
27mCreate Value and Insight Through Privacy-preserving Data Collaboration [...]
01h 10mDIGITAL PLATFORMS, THE NEW PRIVACY CHAMPIONS? BETWEEN MYTHS AND REALITIES
01h 20mUnlocking societal benefits of AI with privacy protective technologies.
32mPrivate Insights into AI Use
21mTowards Building a Responsible Data Economy
28mPrivacy Engineering Networking Break AMA
14mFaster AI model training with confidential computing