Mendel's neuro-symbolic AI beat GPT-4 at retrieving trial-eligible patient cohorts from 1,400 patient records.

Existing systems struggle to accurately identify eligible patients for trials from electronic medical records. Mendel couples language models with a proprietary hypergraph reasoning engine. In the ACR cohort retrieval benchmark with Cornell it outperformed GPT-4 across 1,400 patient records.

Mendel is a clinical AI company whose neuro-symbolic system reads medical records, used for patient cohort matching and trial eligibility

This sits in the Design & Stress-Test stage of the AI use-case lifecycle: Named examples from the tools pharma teams use to scope, prototype and benchmark an AI use case before committing to build it — data-catalog, sandbox and evaluation platforms.