Research & validation
Built for retrospective validation and prospective silent-mode testing.
Echo Pathways is pre-market, and we want it evaluated properly before anyone relies on it. The product is built so that a partner can measure what it would have done — before it does anything at all.
01
Retrospective validation
Replay a historical cohort against a protocol before it goes live. Measure what it catches, when it would have caught it, and — just as importantly — what it misses and how often it fires when it shouldn't.
02
Prospective silent mode
Run live against real results, with alerting switched off. The engine evaluates and records; nobody is paged. You measure real-world performance in your own service without exposing a single patient to an unvalidated system.
03
Real-world evidence
Because every decision is replayable from its recorded inputs, the evaluation dataset is a by-product of running the system — not a separate collection exercise.
Why silent mode matters
Alerting is the last thing we turn on, not the first.
A surveillance system that fires too often is abandoned, and one that fires too rarely is dangerous. Neither can be settled by argument — only by measurement, in the service where it will actually run, against that service's real case mix.
Silent mode exists so that the decision to trust the system is made with evidence, by the clinicians who will carry the consequence. We would rather lose a deployment than win one on a claim we hadn't tested.
Evidence by construction
The audit trail is not a report we write. It is what the system does.
Every data path writes to a processing ledger. Traceability, verification and validation results, the anomaly log and the residual-risk view are emitted by the build rather than hand-authored afterwards — because evidence written after the fact records what somebody believed, not what the system did.
For a research partner, that means the provenance you need for a publication is the same provenance the regulator will ask for, and it already exists.
We're looking for validation partners.
Clinical leaders, digital health teams and research groups with a pathology-driven pathway and a cohort to replay.