Skip to content

VanshAhuja-Auronox/Adaptive-Patient-Monitoring-System

Repository files navigation

Context-Aware Adaptive Patient Monitoring and Risk Escalation System


Project Overview

This project is a command-line based intelligent healthcare system that analyzes patient health data and predicts the risk level using Machine Learning techniques.

In addition to prediction, the system also applies rule-based logic to identify critical conditions and provide appropriate medical recommendations. The goal of this project is to simulate a basic decision-support system that can assist in early health risk detection.


Problem Statement

In real-life situations, patients often struggle to understand the severity of their symptoms based on vital signs such as blood pressure, heart rate, and oxygen level.

There is a need for a system that can:

  • Analyze patient data
  • Predict risk levels
  • Detect emergency conditions
  • Provide simple guidance

Solution Approach

This project uses a hybrid approach:

  • Machine Learning (ML): Predicts risk level based on patient data
  • Rule-Based AI Logic: Refines predictions and detects emergencies
  • Risk Scoring System: Assigns a score to quantify severity

Technologies Used

  • Python
  • pandas
  • scikit-learn
  • pickle

AI & ML Concepts Used

  • Classification (Decision Tree Model)
  • Feature-based prediction
  • Rule-based reasoning (AI)
  • Risk scoring system
  • Basic data preprocessing

Project Structure

  • data_generator.py → Generates dataset
  • dataset.csv → Patient dataset
  • model.py → Trains ML model
  • model.pkl → Saved trained model
  • predict.py → Prediction module
  • decision_logic.py → Rule-based AI system
  • main.py → Main execution file
  • README.md → Project documentation
  • statement.md → Project explanation

Environment Setup

Step 1: Install Python

Make sure Python is installed (version 3.x recommended)

Check using:

  • python --version

Dependencies Installation

Install required libraries:

  • pip install pandas scikit-learn

Configuration Steps

No additional configuration is required.

Ensure all files are in the same folder:

  • dataset.csv
  • model.pkl
  • Python scripts

Execution Steps

Follow these steps in order:


Step 1: Generate Dataset

  • python data_generator.py

Step 2: Train the Model

  • python model.py

This will create:

  • model.pkl

Step 3: Run the System

  • python main.py

Sample Input

Age: 50
Gender: M
Blood Pressure: 150
Heart Rate: 95
SpO2: 92
BMI: 27
Temperature: 101
Smoker: 1
Activity: medium
Duration: 3
Symptom severity: 7


Output Explanation

The system provides:

  • ML Prediction → Initial risk prediction
  • Final Risk → Adjusted risk after AI logic
  • Risk Score (0–100) → Severity measure
  • Advice → Suggested action

Example:

ML Prediction: Medium Final Risk: High Risk Score: 65 Advice: Emergency: Immediate medical attention required


Project Report

The complete project report is available in this repository as Project_Report.pdf.

If the preview is not visible on GitHub, please download the file and open it locally for full content.


System Workflow

  1. User enters patient details
  2. Data is processed and converted
  3. Machine Learning model predicts risk
  4. Rule-based system checks critical conditions
  5. Risk score is calculated
  6. Final output and advice are displayed

Key Features

  • Machine learning-based prediction
  • Rule-based decision system
  • Risk scoring mechanism
  • Emergency detection
  • Simple CLI interface
  • Modular code structure

Use Case

This system can be useful for:

  • Basic health monitoring
  • Educational purposes
  • Understanding AI in healthcare
  • Early risk awareness

Limitations

  • Uses synthetic dataset
  • Not a replacement for medical diagnosis
  • Accuracy depends on data quality

Future Improvements

  • Real hospital dataset integration
  • GUI-based interface
  • Mobile/web application
  • Deep learning models

Author

Vansh Ahuja

VIT Bhopal University
B.Tech — Computer Science & Engineering (Health Informatics)


Interests & Focus Areas

  • Artificial Intelligence in Healthcare
  • Bioinformatics & Medical Data Analysis
  • Data Structures & Data Science

Career Vision

  • To leverage AI and data-driven technologies to revolutionize healthcare systems and improve human life.

Important Note

This project is developed for academic purposes to demonstrate the use of AI and Machine Learning concepts in healthcare systems.

Releases

No releases published

Packages

 
 
 

Contributors

Languages