How Alula's personal model works
Product description. Not medical advice. Not reviewed by a clinician. Updated 24 October 2026.
Alula starts as a PCOS symptom tracker. Over time the logs become a personal model of your record. The app calls that model a digital twin. The practical idea comes first: your days, in one place, with the patterns that actually show up in what you entered.
The twin is not a copy of your body. It does not discover a cause, and it does not try a supplement or medication for you. No clinical validation is claimed.
Three steps
- Track. Symptoms, cycle days, sleep, mood, food, and the supplements or medications you already take.
- Keep the timeline. One day is a note. Many days are a record you can compare.
- Explore patterns. Alula can point at places where something you logged lines up with how you felt later. That is an association in your logs. It is not proof that one thing caused the other.
What the digital twin is
In the app, the twin is a rules-based model. It is not a machine-learning diagnosis. It is built from symptoms, lifestyle, lab results, body measurements, and logs you entered. A signal can show the inputs that contributed to it. The wording in the product is that a pattern comes from what you entered, and that it is not a diagnosis.
Rules and thresholds are fixed in the product. They are not a personal clinician, and they are not hidden as a claim of medical certainty.
What a pattern means
For a symptom you track, Alula can compare days that followed something you logged with days that did not. Examples of those exposures include a food tag, short sleep, a busy day, a missed dose, a cycle phase, or a logged injection. The comparison uses your own check-ins.
The label depends on how much history you have and whether the direction holds:
- Early signal. There is not enough history yet, or the comparison does not clear a stronger bar.
- Possible pattern. The difference is larger, with more days on both sides, and the same direction in both halves of the window.
- Consistent pattern. A higher bar: a larger difference, more days, and the same direction again across the window.
These labels are there so a thin coincidence is not dressed up as a finding. Foods logged together can travel together, so a weaker pattern is checked again without the stronger ones. None of this establishes a cause. A before-and-after average around a supplement or medication is a description of the record, not evidence that the product worked.
Questions about your logs
You can ask a question about your own record. If there is not enough history to say how something would show up for you, Alula says that. An answer is a reading of what you entered. It is not medical advice and not a prediction that a new treatment will work.
Asking a question sends that question and the most relevant logs you typed to a language-model provider, so it can answer and highlight the twin. A day note can be sent so it can be sorted into supplements, food, and symptoms you already track. Apple Health data is left out of those requests. Details are in the privacy policy.
What Alula does not do
- It does not diagnose PCOS or any other condition.
- It does not tell you to start, stop, or change a medication or supplement.
- It does not claim a chart, a percentage, or a sample story on this website is your result. Samples are labeled as samples.
- It does not offer a shared client journal yet. That is part of what we are building for coaches and practitioners, and only if a client chooses to share. It is not open.
- It is not on the App Store yet. There is no price and no opening date on this site.
The waitlist is an email, not a health form.