Case study · Mobile Apps
HealthEZ
A React Native app for patients and doctors — ML-assisted prescriptions for selected diseases and online consultations.
- Category
- Mobile Apps
- Role
- React Native developer
- Team
- Team of 3
- Duration
- Jul 2021 – Jul 2022
- Technologies
01
Overview
HealthEZ is an automated health app with two goals: helping patients get an online prescription for a selected set of common diseases using machine learning, and letting doctors consult with patients online. Patients describe their symptoms in the app, a model hosted on Google Cloud suggests a prescription for supported conditions, and anything outside that scope goes to an online consultation with a doctor.
02
Key features
- Patient and doctor accounts
- Symptom questionnaire for supported diseases
- ML-assisted prescriptions for selected diseases
- Online consultations with doctors
- Prescription and consultation history
Stack
How it’s built
Mobile app
Back end
Machine learning
03
My contribution
- 01
Patient app
Built the React Native screens patients use: sign-up and sign-in, the symptom questionnaire, the generated prescription and a history of past prescriptions and consultations.
- 02
ML prescription flow
Connected the app to the prescription model hosted on Google Cloud: the app sends the patient’s answers, receives a suggested prescription for supported diseases and presents it clearly.
- 03
Doctor side and consultations
Doctor profiles, incoming consultation requests and the online consultation flow between patient and doctor.
- 04
Firebase back end
User authentication and the database for patients, doctors, consultations and prescriptions.
05
Challenges
Challenge
Using machine learning responsibly in a health app.
How I addressed it
The model only covers a selected set of diseases, and anything outside that scope is routed to an online consultation with a doctor instead of an automated answer.
Challenge
Running a model without slowing down the phone.
How I addressed it
The model runs on Google Cloud behind an API, so the app stays light and the model can be updated without releasing a new version of the app.
Challenge
Keeping medical information private.
How I addressed it
Accounts sign in through Firebase Authentication, and each patient’s prescriptions and consultations are tied to their account rather than shared.
06
Outcome
HealthEZ has no public link to share.
- A working app covering both goals: ML-assisted prescriptions for selected diseases and online consultations with doctors.
07
What I learned
“In health software, knowing what the model should not answer matters as much as what it can — clear limits made the automated part trustworthy.”
Next project
Alaaaldeen
A bilingual (English and Arabic) Flutter shopping app for phones, tablets, laptops and smartwatches.
Have an idea? Let’s build it.
If a project like this is on your roadmap, I’d be glad to hear about it.