---
title: "How it works | Hayhurst Medical Technologies"
description: "How HMT's software moves from a new finding to a clinician-approved next step, how each deployment starts as a silent pilot, and how data is protected along the way."
canonical: https://hayhursttech.com/platform/how-it-works/
---

Platform · How it works

# From a sentence in a report to the right specialist.

Catch the patient, route them toward the right care, and learn from what happened. A clinician approves every step, and every deployment starts in silence.

[Request a briefing](/company/contact/)

The loop

## Catch, route, learn.

1. ### Catch  
A concerning finding appears in a report. The software flags it and shows the sentence behind it.
2. ### Route  
The referring clinician sees a suggested next step and qualified specialists nearby, and decides.
3. ### Learn  
Identifiers are removed before the record enters the registry, which helps the software improve over time.
- Clinician approves every step
- Minimum necessary data
- Zero-trust architecture

Working with a cancer program

## Every deployment starts as a silent pilot.

No disruption to anyone's workflow, no patient affected, and no data leaving where it should be.

1. ### Scope  
We agree with a cancer program on what to cover, and on the agreements and oversight each site requires.
2. ### Silent pilot  
The software runs in the background under IRB oversight and is checked against known outcomes. Nothing reaches a clinician or a patient.
3. ### Launch  
Once the results hold up, suggestions go live for referring clinicians, with training and support.

Data handling

## Protected information stays protected.

Patient information is used only inside secured environments covered by agreements with the organizations that hold it, including business associate agreements where HIPAA requires them, and only as long as it is needed to help refer the patient.

Identifiers are cleared before a record is admitted to the registry. The registry helps the software become more accurate over time and, ideally, improve outcomes.

## Keep reading

- ### [Detection](/platform/detection/)  
What the engine reads, what it produces, and the imaging track.
- ### [Navigation](/platform/navigation/)  
The referring clinician’s view, and why suggestions are neutral by design.
- ### [Registry](/platform/registry/)  
The prospective, outcome-linked record the loop builds.