Meteorosis Labs · Amman, Jordan · est. as Gr8iDoc

Precision software, built by a surgeon.

Eighteen products across the clinic, the classroom and the app store — an ophthalmology EMR, seven enterprise systems running inside a national hospital network, a retina academy, and a shelf of bilingual Android apps. All of it designed, coded and shipped by Dr. Shehab Al-Abed — Head of Medical Retina at the Jordanian Royal Medical Services, RETFound Global investigator, and head of the Jordan A-Eye Research Group.

3 Android apps 15 web applications Oculomics research

Meteorosis Labs
18Products shipped
18+Years in medicine
61K+Visitors on Khetmet Maryam
28Retina lectures published

01 — Google Play · Meteorosis Labs

The Android shelf.

A new developer account, a growing catalogue. Each app is built to run entirely on the device, in Arabic and English, with no account required to start using it.

I Will Guess It: Mind Oracle

رح احزرها
Live

A twenty-questions oracle with a Bayesian inference engine that runs entirely offline. Think of any animal, fruit or vegetable; it narrows 118 candidates by information gain and names yours — then opens an illustrated codex of taxonomy, habitat and folklore.

PKG com.retinarealm.twentyquestionsgame   CAT Trivia   RATED 3+

Get it on Google Play

Pregnancy, Week by Week

حملي أسبوعاً بأسبوع
Testing

An Arabic-first companion through all forty weeks — what is developing, what is changing, and what to raise at the next antenatal visit. Written for Arabic speakers rather than translated at them. Currently with closed-testing users.

PKG arabic.weekly.pregnancy   TRACK Closed testing

Coming to Google Play

Battalt — Quit Smoking

بطّلت
In build

A quitting companion built around the milestones that actually keep people going: time since the last cigarette, cravings ridden out, money kept, and the body's own recovery clock ticking forward hour by hour.

PKG ana.battalt.i_quit_smoking   TRACK In development

In the workshop

02 — The web platform

Fifteen applications, four worlds.

Systems that run real clinics, teach real trainees, carry a hospital network's quality data, and sit on the phones of people who will never know a doctor wrote them. Some are open; the rest run behind a login inside the institutions they were built for.

Enterprise systems

07

Built for institutions rather than individuals: systems carrying a national hospital network's quality data, its complaints, its inventory and its ideas. Internal by nature — listed here because they are the largest things the lab has shipped.

SAMA

Smart Advanced Measurements Analysis
In service

One dashboard aggregating quality indicators from every hospital in the network, kingdom-wide, in place of a reporting cycle that used to arrive by spreadsheet and arrive late.

ROLE Creator & architect   SCOPE Network-wide

SAMA-AI

Forecasting & cross-institutional benchmarking
In service

The intelligence layer above SAMA: forecasting where an indicator is heading, flagging anomalies before they become incidents, and benchmarking large institutions against each other on like-for-like terms.

ROLE Creator   LAYER Predictive analytics

Innovation Portal

Idea submission & structured grading
In service

A route for an idea from a peripheral hospital to reach the centre. Staff submit proposals and a systematic grading scheme scores them at each tier, so good thinking is not lost to distance from headquarters.

ROLE Creator   FLOW Peripheral → central

Complaints Management

Logging, routing & tracking
In service

Every complaint given a record, an owner and a clock — routed, escalated and closed somewhere it can be counted, rather than settled informally and never learned from.

ROLE Creator   TRACK Full lifecycle

Electronic Survey & Reporting

Patient satisfaction, network-wide
In service

Retired the paper patient-satisfaction survey across the network. Collection, sentiment scoring and reporting now run themselves, and the results are comparable between sites for the first time.

ROLE Creator   REPLACED Paper forms

IOL Stock Management

Intraocular lens inventory
In service

Intraocular lenses tracked by power and model across theatres, so the lens a surgeon planned the biometry around is on the shelf on the day of the list.

ROLE Creator   DOMAIN Surgical inventory

Virtual Hospital

RMS telemedicine platform
In service

Sole developer and architect. Secure patient data flows fitted to the hospital's existing workflow rather than bolted beside it, and an appointment surface the telemedicine officers can work at speed in.

STACK ASP.NET · SQL   SINCE 2025

— and others still running quietly inside the institutions they were written for.

03 — Research

Reading the body in the retina.

The clinical software pays for itself. The research is the part that outlasts it — international work on retinal foundation models, and a Jordanian group building on data the field keeps leaving out.

Investigator

RETFound Global

Moorfields Eye Hospital · UCL · NUS · CUHK

The first medical AI foundation model built on genuinely global data. RETFound learned to detect retinal and systemic disease from fundus images and was published in Nature in 2023; the global effort now retrains it across more than 100 million colour fundus photographs contributed by partner groups in over 65 countries — so that retinal AI performs as well in Amman as it does in London.

  • 100M+ fundus photographs
  • 65+ countries
  • Foundation model
Founder & Head

Jordan A-Eye Research Group

Amman, Jordan · Oculomics

Jordan's contribution to that effort. The group builds and validates retinal AI on regional data — Middle Eastern populations that are systematically thin on the ground in the datasets these models train on — and works on oculomics: what an image of the retina can say about the rest of the body.

  • Regional datasets
  • Retinal imaging AI
  • Systemic biomarkers
Founding lead

Diabetic Retinopathy Screening

Royal Medical Services · The Fred Hollows Foundation

Diabetic retinopathy is the leading cause of blindness in working-age adults, and it is silent until it is late. Establishing a dedicated screening clinic at the Royal Medical Services, with the Fred Hollows Foundation, to find it in the years when it can still be treated — the clinical end of everything the research is for.

  • Population screening
  • International partnership
  • Preventable blindness

Oculomics

ˌɒk.jʊˈlɒm.ɪks · noun

The retina is the only place in the body where blood vessels and central nervous tissue can be seen directly, without a single incision. Oculomics is the discipline of reading the rest of the body from that view — cardiovascular risk, diabetic damage, kidney function, the earliest signs of neurodegeneration — all of it leaving traces on a photograph that takes seconds to capture and costs almost nothing.

04 — The lab

One desk, two disciplines.

Meteorosis Labs is not a team. It is a retina consultant who kept noticing that the software in front of him was written by people who had never sat in the chair.

Every product here started as a problem in a real clinic — a list that would not sort, a differential too wide to hold in your head, a unit that kept double-booking injections.

It has run on two tracks at the Royal Medical Services for years. One is clinical: retina fellow, senior specialist, consultant, now head of the department, seeing the cases that get referred when nobody else is sure. The other is quieter — the telemedicine platform, the quality dashboard now reporting from hospitals across the kingdom, the portal that carries an idea from a district hospital all the way to the centre. Same person, same week.

The advantage of building your own tools is that the specification never gets lost in translation: the person writing the scheduling logic is the person who will be running late because of it. That is why OptiVision reads like an exam room rather than a database, why the Uveitis Classifier returns a weighted list instead of a verdict, and why every app ships in Arabic and English from the first release rather than the third.

The same instinct runs outside medicine. A Quran circle that needed to be shared without an account. A pregnancy guide that deserved to be written in Arabic, not translated into it. A guessing game that had no business being as well-engineered as it is.

The research runs on the same current. A foundation model trained on a hundred million retinas is only as good as the populations inside it — which is why the Jordan group exists, and why the data it contributes comes out of the same clinics the EMR was written for.

The lab was known as Gr8iDoc for years and still lives at idoc.site. The name changed; the workbench did not.

05 — Connect

Find the lab.

Clinical teaching, product news and the occasional look inside a build — plus a direct line if you want something made.