Welcome to
The Shell

Our custom built Clinical AI safety system

Welcome to
The Shell

Our custom built Clinical AI safety system

Healthcare has no margin for error

But the dangerous truth is, most AI medical scribes make mistakes.
We are dedicated to removing these risks to patient safety.


Interested in reading more detail?

Healthcare has no margin for error

But the dangerous truth is, most AI medical scribes make mistakes.
We are dedicated to removing these risks to patient safety.


Interested in reading more detail?

That’s why we built The Shell

That’s why we built The Shell

That’s why we built
The Shell

TORTUS’ proprietary clinical safety system, designed to meticulously moderate and regulate our services on a constant basis - catching rare errors before they reach clinicians.

Diagram of our process
Diagram of our process

Errors are caught instantly, not discovered later.

Validates each statement in the note against the source context

It catches hallucinations, removing 93.1% of them before the note is even complete.

It catches hallucinations before reaching clinicians

Zero margin for error

Zero margin for error

While others rely on hoping their models improve, we’ve built a system that guarantees clinical safety through continuous validation.
The Shell is a living safety standard that protects every output, every time.

Unrivalled safety

Unrivalled safety

Even a “2% error rate” can be unsafe when every note touches real patients.
The Shell is engineered to close that final safety gap.

We ran a comparative analysis against other clinical documentation AI systems. Using a controlled set of 50 general practice consultations, we generated notes in the standard SOAP format (Subjective, Objective, Assessment, Plan) across multiple platforms.

From the blog

From the blog

For the deeper divers out there, we started writing our blog. Read more on how we reached this amazing level of technology.

Hardening The Shell: Taking AI safety from detection to action

How we built an end-to-end hallucination detection and correction pipeline that automatically removes 93% of detected hallucinations in clinical notes, increasing hallucination-free notes from 34% to 86% in production

Hardening The Shell: Taking AI safety from detection to action

How we built an end-to-end hallucination detection and correction pipeline that automatically removes 93% of detected hallucinations in clinical notes, increasing hallucination-free notes from 34% to 86% in production

Hardening The Shell: Taking AI safety from detection to action

How we built an end-to-end hallucination detection and correction pipeline that automatically removes 93% of detected hallucinations in clinical notes, increasing hallucination-free notes from 34% to 86% in production

Our partners

Our partners

These are the partners that choose to trust us with their clinical workflow.