PulseNet 🩸

AI FOR GOOD 2.0 Hackathon — Blood Warriors Foundation
AI-Enabled Care Coordination & Access for Thalassemia Patients

---

Architecture

┌─────────────────────────────────────────────────────┐
│                    AWS Cloud                        │
│                                                     │
│  ┌──────────────┐    ┌──────────────────────────┐   │
│  │  App Runner  │    │  App Runner              │   │
│  │  (Frontend)  │───▶│  (FastAPI Backend)       │   │
│  │  React+Vite  │    │  Port 8000               │   │
│  └──────────────┘    └────────────┬─────────────┘   │
│                                   │                  │
│  ┌────────────┐  ┌─────────────┐  │  ┌───────────┐  │
│  │ SageMaker  │  │  AWS RDS    │◀─┘  │  Bedrock  │  │
│  │ XGBoost    │  │ PostgreSQL  │     │  Claude   │  │
│  │  Endpoint  │  └─────────────┘     │  Haiku    │  │
│  └────────────┘                      └───────────┘  │
│                                                     │
│  ┌────────────┐  ┌─────────────┐  ┌─────────────┐  │
│  │   Cognito  │  │  SNS / SES  │  │  CloudWatch │  │
│  │    Auth    │  │  Notifs     │  │   Logging   │  │
│  └────────────┘  └─────────────┘  └─────────────┘  │
└─────────────────────────────────────────────────────┘

Team Branch Strategy

Branch	Owner	Focus
main	All	Baseline walking skeleton (this)
feature/donor-flow	Teammate 1	Donor profile management, webhooks, re-engagement
feature/patient-flow	Teammate 2	Patient intake, transfusion calendar, Bridge mapping
feature/admin-ai	Teammate 3	Coordinator dashboard, XGBoost model integration, Bedrock AI

---

Quick Start (Local)

Prerequisites

• Docker Desktop
• Node.js 20+
• Python 3.11+
• AWS CLI (configured with hackathon credentials)

1. Set up environment

cd pulsenet
cp .env.example .env
# Edit .env — add your AWS RDS DATABASE_URL from the hackathon portal

2. Run with Docker Compose

docker compose up --build

Service	URL
Frontend (Vite)	http://localhost:5173
Backend (FastAPI)	http://localhost:8000
API Docs	http://localhost:8000/api/docs

3. Or run services individually

Backend:

cd backend
pip install -r requirements.txt
uvicorn main:app --reload --port 8000

Frontend:

cd frontend
npm install
npm run dev     # → http://localhost:5173

---

API Endpoints

Health

GET /api/health           → { status: "ok", service: "pulsenet-backend" }

Donor

GET  /api/donor/                    → List eligible donors
GET  /api/donor/{id}                → Get donor profile
PUT  /api/donor/{id}                → Update profile
POST /api/donor/webhook/availability → Availability confirmation webhook

Patient

POST /api/patient/register          → Register new patient
GET  /api/patient/{id}              → Get profile
POST /api/patient/{id}/transfusion  → Log completed transfusion
GET  /api/patient/{id}/schedule     → Upcoming transfusion dates

Admin / AI

GET /api/admin/stats                → Dashboard statistics
GET /api/admin/bridge/mock          → 🤖 ML-ranked donor list (E2E demo)
GET /api/admin/bridges              → All active bridges
GET /api/admin/donors/inactive      → Re-engagement targets

---

AWS Deployment

# Make deploy script executable
chmod +x aws/deploy.sh

# Deploy to development
./aws/deploy.sh dev

# Deploy to production
./aws/deploy.sh production

Plugging in the XGBoost Model

1. Train your model and export it: joblib.dump(model, 'xgboost_model.pkl')
2. Drop xgboost_model.pkl into backend/services/
3. The rank_donors() function auto-detects and uses it — no code changes needed
4. For SageMaker endpoint, update SAGEMAKER_ENDPOINT_NAME in .env

Plugging in Amazon Bedrock

Add to backend/services/bedrock.py using the pattern:

import boto3
bedrock = boto3.client('bedrock-runtime', region_name='us-east-1')
response = bedrock.invoke_model(modelId='anthropic.claude-3-haiku-20240307-v1:0', ...)

---

Dataset → Schema Mapping

Dataset Column	DB Column	Model
user_id	external_id	User
blood_group	blood_group	User
eligibility_status	eligibility_status	User
calls_to_donations_ratio	calls_to_donations_ratio	User
expected_next_transfusion_date	expected_next_transfusion_date	Bridge
bridge_id	external_bridge_id	Bridge
donated_earlier	donated_earlier	BridgeMember
user_donation_active_status	user_donation_active_status	User
inactive_trigger_comment	inactive_trigger_comment	User

---

Tech Stack

Layer	Technology
Frontend	React 18 + Vite + TypeScript + Tailwind CSS
Backend	FastAPI + Python 3.11 (async)
ORM	SQLAlchemy 2.0 (async) + Pydantic v2
Database	AWS RDS (PostgreSQL 15)
ML	XGBoost + SageMaker
AI	Amazon Bedrock (Claude Haiku)
Auth	AWS Cognito
Notifications	AWS SNS + SES
Deploy	AWS App Runner + ECR
Monitoring	AWS CloudWatch
Container	Docker + Nginx