AId Centric builds AI that turns conversations, records, and production data into structured drafts. Every line links to its source, and a named person reviews it before it counts.
Products
Five clinical products and one operational product, shown one after another.
Clinical
Operational
Listening
Conversation
Consented consultation, transcribed
What brings you in today?
This cough has been going on about three weeks. It’s worse at night.
Any fever, or are you bringing anything up?
No fever. A little clear phlegm in the mornings.
I started the metformin last month, the five hundred one.
Your chest sounds clear. If it isn’t better in two weeks, we’ll get a chest X-ray.
Structured draft
Draft note
Select a draft line to see the sources it came from.
Listening
Conversation
Spoken family history
I’m 38, and I’m the one being seen today.
My mum had breast cancer in her forties. Dad’s 67 and well.
I’ve got a younger brother, he’s fine.
My aunt had it too. She died at 52.
Is this aunt your mother’s sister or your father’s sister? Answer: mother’s sister.
Structured draft
Pedigree, standard nomenclature
Family members
Select a draft line to see the sources it came from.
Patient summary
64-year-old, rectal adenocarcinoma, first MDT review
Synthetic patient, colorectal tumour board
Documents
Department agents
Each agent reads the case documents and prepares its department’s considerations.
Gastroenterology agent
ReadyD1
Pathology agent
ReadyD2
Radiology agent
ReadyD3D4
Colorectal surgery agent
ReadyD3D1
General surgery agent
ReadyD5
Genetics agent
ReadyD2
Medical oncology agent
ReadyD2D3D5
Board preparation
Prepared by AI agents for discussion. Not a recommendation.
Agreement
Locally advanced mid-rectal cancer on MRI. Colorectal surgery and medical oncology both raise neoadjuvant treatment before surgery.
Divergence
Colorectal surgery and medical oncology raise treatment sequencing differently. This is left open for the board to discuss.
Missing before decision
Upcoming meetings
Oct1
13:00 – 14:00Weekly on ThuColorectal
Colorectal tumour board
7 cases prepared, 2 with missing items.
Patients7
Invitees9
Main campusRoom 2.14 (2nd floor)
Oct5
12:00 – 13:00Weekly on MonUpper GI
Upper GI tumour board
4 cases prepared.
Patients4
Invitees8
Oncology buildingConference room 4
Oct7
10:00 – 11:30Every 2 weeks on WedHepatobiliary
Hepatobiliary tumour board
No cases added yet.
Patients0
Invitees7
Main campusRoom 6.04 (6th floor)
Past meetings
Sep24
13:00 – 14:00Colorectal
Colorectal tumour board
Patients7
Invitees9
Tracking
Camera view
Lateral shoulder raise
Set 1 of 2Target 75–90°
Shoulder angle
83°
Reps
6/6
Set complete. 4 of 6 reps in range.
Computed in code, not by a language model
Study question
Which factors predict ICU admission within 48 hours of an emergency admission?
1. Data source
Done
SELECT a.admission_id, p.age, p.sex,
v.resp_rate, v.spo2, l.lactate,
l.crp, l.wbc, a.icu_within_48h
FROM ed_admissions a
JOIN patients p USING (patient_id)
LEFT JOIN first_vitals v
USING (admission_id)
LEFT JOIN first_labs l
USING (admission_id)
WHERE a.admitted_at >= '2024-01-01'
AND a.admitted_at < '2026-01-01';ed_admissions_2024_2025.csv
2. Preprocessing
Done3. Missing data
DoneMissing values by variable
Multiple imputation by chained equations (MICE), 5 datasets. The method and seed are recorded with the results.
BMI is 22% missing. Check whether it should stay in the model.
4. Analysis and models
Done5. Analytics dashboard
DoneGradient boosting AUC 0.84Logistic regression AUC 0.79
| ICU | No ICU | p | |
|---|---|---|---|
| Age, median | 71 | 63 | <0.001 |
| Respiratory rate | 24 | 18 | <0.001 |
| Lactate, mmol/L | 2.8 | 1.4 | <0.001 |
| Female | 44% | 51% | 0.002 |
Numbers come from the database, not from the language model.
Which products had abnormal scrap rates this week?
How the answer was built
Question understood
Scrap rate this week, compared with each product’s own baseline
Schema read
quality.weekly_scrap_vs_baseline, production.products
SQL generated
SELECT product_code, plant, scrap_rate,
baseline_12w, z_score
FROM quality.weekly_scrap_vs_baseline
WHERE week = '2026-W37'
AND plant IN ('A', 'B', 'C')
AND z_score >= 3
ORDER BY z_score DESC;Validated
Scope and permissions checked before running
Ran on the database
0.4 s, 214 products checked, 3 rows
Result
The answer appears here once the query has run.
3 of 214 products are 3σ or more above their 12-week baseline. PX-2210 in Plant B has the largest deviation.
3σ band around the 12-week baselineBaseline
Flagged products
Select a product to see its chart.
Where it helps
The clinician keeps their attention on the patient. The note is drafted for review.
A patient follows the programme their physiotherapist set, and the physiotherapist sees how it went.
A planner asks a question and can check the query behind every answer.
Evidence
External studies
of physicians’ office day went to EHR and desk work in a time and motion study of 57 US physicians. Only 27% went to direct face time with patients.
Sinsky et al., Ann Intern Med 2016physicians took part in a randomised trial of two ambient AI scribes. One reduced time in notes by 9.5% against control, the other showed no significant change, and occasional inaccuracies were reported for both.
Lukac et al., NEJM AI 2025studies with 1,520 participants in a meta-analysis found real-time telerehabilitation comparable to in-person care for physical function in musculoskeletal conditions.
Cottrell et al., Clin Rehabil 2017We start with synthetic data. Real data comes only after boundaries and approvals are agreed.