Presented at "21st BioPharma Clinical Trials Nexus Conference, New Jersey" 20241023_1115

presentation.pdf

presentation.mp4

Using AI to Traverse the Path from Clinical Trials to Personalized Medicine

• Rare Cancers: I have one. They aren’t researched much. How I deal with it.
• Oncologic Clinical Trials: Success rate is lousy. My goal is to increase it.
• Portfolios, Juries, Parole and Tumor Boards: I do this using Personalized Tumor Boards.
  • Tumor Boards:
    • “A meeting of healthcare professionals that discuss complex cancer cases to determine the best treatment plan for a patient.”
  • Insights from two 13 member Tumor Boards I created.
    • EVEREST: "Should Bill participate in the EVEREST Trial".
    • Pro/Photon: "Should Bill use Proton or Photon radiation to treat his Brain Tumor."
• Hackathons: 80 person competition between 17 “gamified” Tumor Boards
  • "Result ranking" functions are used to determine who won the game.
• LLMs: Increase tumor board membership to 15 by adding claude.ai and chatGPT-4o.
• Moderators: “Moderate” the EVEREST and Pro/Photon Tumor Boards using LLMs.
  • Human tumor board members spend a lot of time explaining their positions to each other. 
• You can help!.