Personal Health Knowledge Graph
A longitudinal timeline per patient, resolved against a standards ontology backbone.
SNOMED-CTLOINCRxNormICD-10CPT
The platform
One engine between the record and the decision
A patient's history arrives in four incompatible shapes. The Health Context Engine resolves them into a single longitudinal graph, reasons over it against clinical evidence, and returns answers with citations attached.
Benchmark88% endocrinology
Hallucination rate0.8%
Ontologies5 standards
01Architecture
Three layers, one contract
Nothing here is a black box. Each layer has a defined input, a defined output, and a citation trail that survives clinical review.
02Why it holds up
Grounded, traceable, deterministic
General-purpose models improvise when the record is thin. A clinical system cannot. January's engine pairs retrieval with deterministic rules so the same question returns the same answer, with the same sources.
88%
Highest endocrinology benchmark score
Our Health Context Engine outperforms GPT, Gemini, and human clinicians on the endocrine medical board exam. Its proprietary RAG reaches higher accuracy without web access to guidelines and literature.
Endocrine Medical Board Exam results; January AI RAG performance vs. GPT, Gemini, and human users
0.8%
Hallucination rate
Against 1.8% contradictions, materially below general-purpose frontier models on the same clinical cases (−log₁₀(p) = 8.0 and 8.4).
Internal evaluation vs. GPT and Gemini across clinical case set
03Integration
Meet your stack where it is
Easy-to-integrate APIs
REST endpoints for glucose prediction, health context, and nutrition, backed by our 54M verified-food database. Ship in days, not quarters.
EHR-connected workflows
Epic-integrated and qualified on Mayo Clinic Platform, insights land where clinicians already work.
White-label applications
Deploy the full consumer experience under your own brand, on your own terms.
Get started