Product

Raw data in. Routed, explainable context out.

Thalamus is a four-layer pipeline that turns every document a patient generates into a live, queryable patient state — and compiles the right slice of it for each agent, each time.

L1 Raw data L2 Atomic facts L3 Patient state L4 Agent context
L1

Raw clinical data

Every source document a patient generates, kept in full. Nothing is discarded here — this layer is the system of record underneath everything Thalamus computes.

L2

Atomic Clinical Data Points

Every document is broken down into discrete clinical facts — Atomic Clinical Data Points, or ACDPs. Each one carries roughly ten metadata tags today, a set built to grow past a hundred: category, organ system, asserter, temporal information, source and confidence among them. A single patient can hold thousands.

Diagnosed with dyslipidemia
LDL = 168 mg/dL
Underwent bilateral RIRS
Complains of right flank pain
Prescribed Atorvastatin
Creatinine = 1.1

Every ACDP links back to its original document, timestamp and provenance.

L3

Computed patient state

L3 performs inference over the atomic layer to determine the patient's current state — and stays current automatically as new documents arrive.

L4

The intelligence layer

Three capabilities sit on top of the computed patient state, each consuming it differently.

L4A

Agent context compiler

Different agents need different context — a pregnancy agent doesn't need a full dermatology history. The compiler decides which patient state, which ACDPs and which context each specific agent receives, and prepares the input it's given.

L4B

Guideline recommender

Ingests clinical guidelines from bodies including ADA, USPSTF, AHA, ASCO and ATA, and determines which apply, which recommendations are relevant, and which screenings should surface for this patient's current state.

L4C

Clinical intelligence

Proactively surfaces care gaps, preventive opportunities, missing follow-ups, clinical risks and unresolved issues — before the downstream agent starts reasoning.

Continuous updating

Event-driven, not batch.

Every new document — a lab result, a follow-up note, a five-minute agent call — flows through the same pipeline the moment it arrives, so the patient profile never goes stale.

New document L1 — ingested L2 — ACDP extraction L3 — patient state recomputed L4 — intelligence updates automatically
Explainability

Every answer traces back to a source.

See it against your own data.

Bring a real patient record set and we'll walk through what Thalamus routes to your agent.

Book a demo