Kardiosense

Kardiosense is an AI-powered cardiac risk prediction platform built in Uganda for clinicians who need reliable, offline-aware tools for cardiac screening and follow-up.

Mission statement: Taking care of your heart, beat by beat

Repository Structure

Path	Purpose
.github/workflows/	CI pipelines for mobile, backend, and ML workstreams.
mobile/	Flutter Android application for clinicians.
backend/	FastAPI service, API routes, schemas, services, Docker setup, and Alembic config.
ml/	AI/ML pipeline for ECG, clinical, fusion, and TensorFlow Lite export work.
supabase/	Supabase SQL migrations and seed data.
docs/	Product, architecture, UX, schema, and implementation documentation.

Prerequisites

• Flutter 3.19+
• Python 3.11+
• Docker
• Supabase CLI

Quickstart

Create your local environment file:

cp .env.example .env

Mobile workstream:

cd mobile
flutter pub get
flutter run --dart-define=SUPABASE_URL="$SUPABASE_URL" --dart-define=SUPABASE_ANON_KEY="$SUPABASE_ANON_KEY"

Backend workstream:

cd backend
python -m venv .venv
.\.venv\Scripts\Activate.ps1
pip install -r requirements.txt
uvicorn app.main:app --reload

ML workstream:

cd ml
python -m venv .venv
.\.venv\Scripts\Activate.ps1
pip install -r requirements.txt
python -m pytest

Supabase workstream:

supabase start
supabase db reset

Docker backend:

docker compose up --build

The root Compose file starts the FastAPI backend on http://127.0.0.1:8000 and Redis on localhost:6379. Keep Flutter mobile builds on the host machine so the app can use the local Flutter SDK, emulators, desktop build tools, and connected devices directly.