Regulatory Position

The short version

Medi-Chi is built as practitioner-facing clinical decision support. Every clinical output is advisory, is directed at a qualified practitioner rather than the public, and is designed so that you can independently review the reasoning behind it instead of having to take it on trust.

Nothing here acts on its own output. No recommendation reaches a patient without a registered practitioner reading it, judging it, and deciding.

What 'independently reviewable' means here

The design goal is that you never have to rely on the software. For any clinical output, you should be able to see:

The information the software relied on
Why a pattern was selected
Which competing patterns were considered, and the evidence against each
The classical and research sources behind the reasoning
Contraindications, red flags and restricted-substance warnings
How confident the output is, and where the data was too sparse to be confident
A clear path to reject or modify anything before it reaches a patient

Where a function does not yet meet all of these, that is recorded as an open item in our internal classification rather than glossed over.

Function by function

Diagnosis engine

Intended purpose: Organises practitioner-entered findings into candidate TCM pattern differentiations with points, formula options and cited reasoning.
What it is not: It does not diagnose a patient, issue a treatment order, or act on its own output.
How you review it: Three ranked candidates with a confidence score, evidence for and against each, and named classical sources.

Tongue image analysis

Intended purpose: Describes observable features of a practitioner-captured tongue photograph and relates them to candidate patterns.
What it is not: It does not screen for, detect or exclude any biomedical disease, oral pathology or malignancy from an image.
How you review it: Descriptive observations with confidence bands, checked against your own in-person examination. Photographic colour is unreliable — your examination governs.

Reverse diagnosis

Intended purpose: Suggests Western biomedical terminology that may correspond to a pattern picture you have already formed, for referral letters and notes.
What it is not: It is terminology mapping for communication. It is not a biomedical differential diagnosis.
How you review it: Multiple candidate terms with the reasoning for each.

Quick questionnaire

Intended purpose: An educational orientation aid that maps questionnaire responses to a provisional pattern shortlist.
What it is not: It does not diagnose, and it does not use model inference.
How you review it: The logic is a fixed, deterministic rule set — the basis is the rules themselves.

Reports and handouts

Intended purpose: Formats clinical documents from content you have entered, reviewed and approved.
What it is not: It generates no clinical recommendation of its own.
How you review it: You edit every document before it is issued.

Patient records

Intended purpose: Stores and retrieves your clinical records under access control.
What it is not: No inference, scoring or alerting is derived from stored records.
How you review it: Not applicable — no clinical conclusion is produced.

Tongue image analysis is held to a higher bar

Software that reads an image and produces a clinically relevant conclusion sits in a different risk class to text-based support, and we treat it that way. A practitioner can re-read a symptom list; nobody can easily re-derive what a model saw in a photograph.

So: the tongue analysis describes the photograph you uploaded, nothing more. Photographic colour rendition is unreliable under clinic lighting, and your in-person examination always governs. It does not look for oral pathology, malignancy or systemic disease, and it must never be used to reassure anyone that a visible lesion is benign. Anything on a tongue that looks wrong to you clinically goes to medical referral, not into a pattern.

What we are not claiming

This page describes how the software is designed and intended to be used. It is not legal advice, and it is not a determination of regulatory status.

Medi-Chi has not:

been assessed or registered as a medical device under the TGA, FDA, EU MDR or UK MHRA frameworks;
been validated in a prospective clinical trial — we make no accuracy or outcome claims;
been certified against any compliance or security standard.

If any of that changes, it will be recorded here with the date and system version.

Where the responsibility sits

The final diagnostic and treatment decision rests solely with the registered practitioner. Medi-Chi does not diagnose, treat, cure or prevent disease, and its output is not medical advice to the general public. As the treating practitioner you remain responsible for consent, records, scope of practice and your own regulator's requirements.