This project implements an AI-powered triage system designed to streamline the triage process in emergency rooms by prioritizing patients based on the severity of their condition using machine learning techniques.
- Patient Data Management: Load, preprocess, and manage patient data from a CSV file.
- Triage Classification: The system determines a patient’s triage level based on their symptoms and vitals.
- Machine Learning: Utilizes a Random Forest Classifier to predict triage levels based on historical data.
- Real-Time Data Integration: Simulates real-time patient data and integrates it into the triage queue.
- Graphical User Interface (GUI): Built with Tkinter to allow for easy data entry, display the patient queue, and visualize triage distribution.
- Data Security: Incorporates encryption and decryption of sensitive patient data using cryptography's Fernet module.
Install the necessary Python libraries using: bash pip install -r requirements.txt
requirements.txt includes:
- pandas
- numpy
- scikit-learn
- tkinter
- matplotlib
- seaborn
- cryptography
- Load Patient Data: The patient data is loaded from a CSV file.
- Preprocess the Data: The data undergoes preprocessing, which includes imputation of missing values, scaling of numeric data, and encoding of categorical variables.
- Model Training: The system trains a Random Forest Classifier to predict triage levels.
- GUI: The user can input new patient data via the GUI and view real-time updates of the triage queue and distribution graph.
To run the program: bash python main.py
The GUI enables the user to:
- Enter patient details like age, gender, symptoms, and vitals.
- View the patient queue sorted by triage level.
- See a graphical representation of the current triage level distribution.
Patient data is encrypted using the Fernet encryption system to ensure confidentiality.
This structure includes essential project details, setup instructions, and how to run the system. Let me know if you need any specific tweaks!