×
Open-source AI speeds up patient-matching for clinical trials
Written by
Published on
Join our daily newsletter for breaking news, product launches and deals, research breakdowns, and other industry-leading AI coverage
Join Now

The doctor will see you right now…

The healthcare industry has long struggled with efficiently matching patients to clinical trials, with traditional methods taking hundreds of days and resulting in 80% of trials missing enrollment targets. Mendel AI, a clinical AI platform, is addressing this challenge by combining open source AI models with specialized healthcare technology to reduce matching time to just one day.

The innovation breakthrough: Mendel AI’s Hypercube platform integrates Meta’s open source Llama model with a clinical hypergraph to revolutionize patient matching and clinical trial management.

  • The platform enables healthcare companies to organize data on their own cloud infrastructure, creating a secure and searchable knowledge base
  • Hypercube supports natural language querying for trial matching and patient cohort analysis
  • The system handles complex clinical tasks including data abstraction and chart reviews

Technical implementation: Mendel AI has developed a sophisticated approach to customizing open source AI for healthcare applications.

  • The company began by fine-tuning Llama 2 to create a natural language interface for their knowledge base inference engine
  • Engineers continuously pre-trained both 8B and 70B parameter versions of Llama 3 to develop healthcare-specific foundation models
  • The resulting system combines lightweight models with instruction-following and in-context learning capabilities in an agentic framework

Strategic advantages: The use of open source AI models provides significant benefits for companies operating in regulated industries like healthcare.

  • Organizations can create customized AI solutions without sharing sensitive data with model providers
  • The approach eliminates substantial upfront costs associated with proprietary AI systems
  • Companies can focus resources on developing industry-specific applications rather than basic AI infrastructure

Expert perspective: Dr. Wael Salloum, Mendel’s Founder and Chief Science Officer, emphasizes the transformative potential of their approach.

  • The platform enables rapid innovation by allowing companies to focus on specific use cases
  • Open source models democratize access to advanced AI capabilities for startups
  • Mendel AI plans to incorporate the multimodal Llama 3.2 model in future developments

Future implications: The successful implementation of open source AI in clinical trial matching could mark a turning point in healthcare innovation, though questions remain about scalability and regulatory compliance across different healthcare systems and jurisdictions.

Dramatically accelerating patient-matching in clinical trials with open source

Recent News

Sam Houston State to launch new AI and IT college

The Texas university will offer online technical certificates and associate degrees in AI, cybersecurity and paralegal studies starting fall 2025.

AI search startup Genspark secures $100M to challenge Google

New search startup raises $100M to provide AI-verified answers that eliminate the need to click through multiple web links.

Go small or go home: SLMs outperform LLMs with test-time scaling

Small models achieve GPT-4-level performance on specific tasks through smarter optimization techniques, using a fraction of the computing power.