Patient message
Saved before anything else happens, so nothing is lost if a later step fails.
How it works
Patients check in with Imori between visits through a short, plain-language conversation, by voice, chat or text. Every message is scored for risk against named clinical categories. The most important messages are elevated in your queue, with the excerpt and the reasoning. Every interaction is summarized, and every message stays available for you to review.
Saved before anything else happens, so nothing is lost if a later step fails.
Scored 1–10 against named categories, with factors, reasoning and the quoted excerpt.
Messages at or above your threshold are elevated, grouped by why. Your protocol decides who is notified and how.
A clinician reviews, and the record keeps who did so and when. The interaction, the summary and the time spent are documented for you.
One patient · 6 weeks · 42 check-in messages
Every message is assessed against named categories — ideation and self-harm, hopelessness, substance use, psychosis, severe depressive symptoms, isolation, functional impairment, crisis language.
Illustrative — not patient data
The two sides
Imori is two screens that share one record. The patient has a conversation. You have a queue.
Signs up on the web or in the app and completes a short intake. The questions and instruments are the ones your clinic chooses; PHQ-9 and GAD-7 are common examples.
After that, Imori is a plain conversation they open when they want to, or reply to by voice or text. It asks how the week is going, how they are sleeping, what is on their mind, and it answers in kind.
When you request an instrument such as a PHQ-9 or GAD-7, it arrives inside that same conversation with a consent prompt first. The patient can say not now.
Where your clinic wants it, the crisis line appears on every screen, and the conversation points to it, or to emergency services, when a message calls for that. This is configurable, and it carries your clinic's name and brand.
Where your clinic supports it, a patient can ask to schedule a conversation or be connected to a telehealth call, and a care provider can request one too. This is optional, and it depends on your availability.
The patient never sees a score, a flag, or the word risk. Their side is a conversation, and it stays one.
Every patient message is scored in the background, moments after it is sent, against named clinical categories. The score, the factors, the reasoning, and any immediate concern are stored with the message that produced them.
Your dashboard opens on the queue. Patients are grouped by why they are there: crisis language, high risk and unreviewed, a worsening score on an instrument, a patient who has stopped responding, and other groups your clinic sets.
Open one and you see the excerpt in context, the score, the factors, the reasoning, the care team, the latest instruments with their trend, and any notes.
You mark it reviewed, request an instrument, schedule a conversation, start a telehealth call, or add a note. Marking reviewed records who did it and when.
Notifications follow your protocol. Nothing is decided for you.
The summary and the time spent are written for you, ready to review and sign. Every message stays available to read.
Documentation
Each patient interaction becomes a structured record: the conversation, a short summary, any instrument scores, and the time your team spent reviewing and coordinating. Summaries are aligned to the care management codes your clinic works under. Your clinicians review and sign, and you bill through your own process.
Imori's web interface is where your team works. Integration with your clinic's EHR is built around your workflow once a pilot has shown its value.
Before you ask
What does the scoring?
A large language model, run as a separate pass after every patient message. It reads the last few turns and returns a score from 1 to 10, the factors it saw, its reasoning, and any immediate concern, as structured data you can inspect. There is no keyword list.
Where is the line?
Seven and above goes to the queue by default, and the threshold is configurable for your clinic. Scores below it stay with the conversation, summarized and there when you want to review.
What happens in a crisis?
This is one example of what Imori can do. The patient moves to the top of the queue, and your protocol takes over: the people you named are notified, in the order and by the means you chose, and, if your protocol says so, emergency services are contacted. If your clinic chooses, the conversation also points the patient to the crisis line.
What if nobody is on call?
Imori is available around the clock. It stays with the patient and notifies your care providers so they can respond as soon as they are available.
Which channels?
Voice, chat, text and the Imori app. Patients choose what works for them, and your clinic chooses which to offer.
Which instruments?
PHQ-9 and GAD-7 are common starting points, and AUDIT is available for alcohol use. These are examples. We can implement whatever tools your clinic wants, or whatever makes sense for your workflow, scored by the published rules and charted over time.
Does it replace my team?
No. It makes your team more efficient and your care more effective, and it helps ensure your clinic is compensated for the work being done. Your clinicians stay in charge of every decision.
Boundaries
Imori does not replace your clinical team. Imori keeps clinical judgment with the clinician.
It does not diagnose, and it does not suggest a diagnosis.
It does not deliver therapy, and it does not purport to be a therapist.
It does not act outside a protocol your clinicians write and approve.
It does not replace your screening instruments. It delivers the ones you choose into the conversation.
It does not bill. You bill through your own process.
It is not a crisis line, but it can connect the patient with the number anytime it matters.
Let's talk, and we will walk you through the queue on sample records.
Contact us