Feb 1, 2026

SmartCare

healthcare health smartcare healthtech ai

SmartCare – Complete Project Documentation

AI-Powered Patient Queue and Triage Optimization System
Built for HackNagpur Hackathon 2026


Table of Contents

  1. Project Overview

  2. How It Works

  3. System Features

  4. Technical Architecture

  5. AI / ML Components

  6. User Roles and Credentials

  7. API Endpoints

  8. Installation Guide

  9. Deployment Guide

  10. Demo Walkthrough


1. Project Overview

Problem Statement

Traditional healthcare facilities face several operational challenges:

  • Long patient wait times, especially in emergency departments

  • First-come-first-served queues that ignore medical urgency

  • Overloaded doctors and uneven workload distribution

  • Manual triage processes that delay critical care

Proposed Solution

SmartCare is an intelligent healthcare management system that uses AI-assisted triage to prioritize patients based on medical urgency rather than arrival time.

Flow:
Patient Check-in → AI Symptom Analysis → Priority Score (1–10) → Smart Queue → Doctor Assignment

Key Benefits

  • Up to 40 percent reduction in average waiting time

  • 60 percent faster response for critical cases

  • Balanced doctor workload

  • Improved patient satisfaction and safety


2. How It Works

System Workflow

  1. Patient Check-in
    Patients enter symptoms, age, and basic medical details through a web interface.

  2. AI Triage Engine
    The system analyzes symptoms using rule-based NLP, detects red-flag conditions, considers age and risk factors, and generates a severity score from 1 to 10 with an explanation.

  3. Priority Queue Insertion
    Patients are inserted into a dynamic priority queue based on severity rather than arrival time. Estimated wait time is calculated dynamically.

  4. Smart Doctor Allocation
    Patients are matched to doctors based on department, availability, and workload. A spare doctor pool handles overflow situations.

  5. Consultation
    Doctors consult patients, can override or escalate priority if needed, and mark consultations as completed.


3. System Features

Patient Features

  • Simple symptom-based check-in

  • Fair priority assignment based on urgency

  • Real-time queue position updates

  • Mobile-friendly interface

Doctor Features

  • Dashboard displaying assigned patients

  • Emergency priority override option

  • Access to patient triage details

  • Easy consultation completion workflow

Administrator Features

  • Real-time analytics and queue monitoring

  • Doctor pool management

  • System configuration controls

  • Department-wise statistics


4. Technical Architecture

Technology Stack

Frontend:

  • React 18

  • TypeScript

  • Vite

  • Tailwind CSS

Backend:

  • FastAPI (Python 3.11)

AI / ML:

  • Rule-based NLP

  • Priority scoring algorithms

Database:

  • In-memory storage for demo

  • PostgreSQL for production

Deployment:

  • Docker

  • Railway or Render


5. AI / ML Components

Triage Engine

The triage engine evaluates symptoms using predefined medical keywords and risk factors.

Critical keyword examples:

  • chest pain

  • stroke

  • unconscious

  • severe bleeding

Urgent keyword examples:

  • difficulty breathing

  • high fever

  • severe pain

Additional factors:

  • Age (children under 5 and adults over 65 get higher priority)

  • Duration of symptoms

  • Pre-existing conditions

Sample output:

  • Severity score

  • Severity level

  • Recommended department

  • Explanation


Priority Queue System

A max-heap based priority queue ensures the highest priority patient is always served first.

Priority formula:
(severity × 10) + age factor + wait time bonus

This prevents starvation and ensures fairness.


Doctor Allocation Engine

Doctor assignment considers:

  • Department matching

  • Current workload

  • Availability

  • Spare doctor pool for overflow


6. User Roles and Credentials

Default Credentials

Admin
Email: admin@smartcare.com
Password: admin123

Doctor
Email: doctor@smartcare.com
Password: doctor123

Patients

patient@smartcare.com patient123


Role Permissions

Patients:

  • Check-in

  • View queue

Doctors:

  • View queue

  • Access dashboard

  • Emergency override

Admins:

  • Full system access

  • Analytics

  • Doctor management


7. API Endpoints

Base URL:
http://localhost:8000/api/v1

API Documentation:
http://localhost:8000/docs

Authentication

POST /auth/login
POST /auth/register
GET /auth/me

Smart Queue

POST /smartqueue/checkin
GET /smartqueue/list
GET /smartqueue/status
GET /smartqueue/stats
POST /smartqueue/call-next
POST /smartqueue/complete/{id}
POST /smartqueue/emergency/escalate

Doctors

GET /doctors
GET /doctors/available
POST /doctors


8. Installation Guide

Prerequisites

  • Docker and Docker Compose

  • Node.js 18 or higher

  • Python 3.11 or higher


Docker Setup

 
git clone https://github.com/KomalLaddha-Dev/DevOcs_HackNagpur.git cd DevOcs_HackNagpur docker-compose up --build

Frontend:
http://localhost:3000

Backend:
http://localhost:8000


Manual Setup

Backend:

 
cd backend python -m venv venv venv\Scripts\activate pip install -r requirements.txt uvicorn app.main:app --reload --port 8000

Frontend:

 
cd frontend npm install npm run dev

9. Deployment Guide

Backend and frontend can be deployed separately using Railway or Render.

Environment variables:

Backend:
SECRET_KEY
DEBUG
CORS_ALLOW_ALL

Frontend:
VITE_API_URL


10. Demo Walkthrough

Scenario:
A busy hospital morning with multiple patient arrivals.

  • A low-priority patient with mild symptoms is placed lower in the queue.

  • A critical patient with chest pain arrives later but is immediately prioritized.

  • Doctors always receive the most urgent patient first.

  • Emergency override allows doctors to escalate worsening cases.


Project Repository

GitHub:
https://github.com/KomalLaddha-Dev/DevOcs_HackNagpur


Built for HackNagpur Hackathon 2026

This build was uploaded as a hackathon project

Hackathon

Hacknagpur 2.0

View All Projects
Give a star to encourage!Discussion
Start a new conversation!
Login to join the discussion