Skip to content

Latest commit

 

History

History
74 lines (47 loc) · 4.3 KB

README.md

File metadata and controls

74 lines (47 loc) · 4.3 KB

Finding the Diseases Based on the Symptoms

Overview

The Finding the disease based on symptoms project is an innovative and crucial application of artificial intelligence in the healthcare domain. The primary objective of this project is to provide users with a powerful tool that can assist in identifying potential diseases or health conditions based on the symptoms they report. Leveraging advanced machine learning algorithms and natural language processing techniques, this project aims to offer a preliminary screening system that encourages users to seek professional medical advice promptly.

Upgrades (0.2) - 17 September 2023

New upgrades are now available. What has changed?

  • Supported with Streamlit Library. The old version was running with Flask
  • TF-IDF and HuggingFace Answer-Question system is now available
  • More professional code design so you can run the app.py within 8.2203 seconds
  • An adjustable Patient Profile was added as a side selection
  • Symptom Weight Visualizer was added. It is going to tell you how a symptom can affect your body within two days.

    • How to run the project?

  • Write pip install requirements.txt to CMD(or relevant)
  • You should write streamlit run app.py to CMD(or relevant)
    • If you have a problem doguhannilt@gmail.com

    Dataset

    We already used multiple datasets from Kaggle, you can check the Dataset folder.

    Models and Accuracy

    In this project, we have employed the K-Nearest Neighbors (KNN) Classifier for modeling the disease diagnosis system. KNN is a well-known machine learning algorithm used for classification tasks. The accuracy achieved by our KNN model is an impressive 72% to %85, indicating its potential to make reasonably accurate predictions based on symptom inputs.

    Benefits and Limitations

    This project offers several significant benefits:
  • Accessibility: Users can access the system from the comfort of their homes or on-the-go, enabling faster and more convenient preliminary assessments of their health status.
  • Early Awareness: By encouraging users to be proactive about their health, the system promotes early detection of potential health issues, leading to timely medical intervention.
  • Public Health Impact: Prompt medical attention, facilitated by the system, has the potential to reduce the severity of certain illnesses and improve overall public health outcomes.

    • However, it's crucial to acknowledge the limitations:
  • Not a Replacement for Medical Professionals: The system does not substitute the expertise of medical professionals. Its results should be used as informative guidelines rather than definitive diagnoses.
  • Accuracy: While the KNN model aims to provide accurate predictions, its reliability may vary based on the quality and quantity of data it has been trained on.

    • Checkpoints File

    The checkpoints file contains the Python scripts and HTML/CSS codes that have been written to develop the disease diagnosis system. You are welcome to use these resources as needed!

    Disclaimer

    It's essential to remember that this disease diagnosis system is not a substitute for professional medical advice. Always consult qualified healthcare professionals for a comprehensive evaluation and proper medical care. This project is designed to complement medical expertise and empower users to take an active role in their healthcare journey.

    We hope that this disease diagnosis project can be a valuable tool in promoting early detection and improving healthcare accessibility. Your feedback and contributions are welcomed to further enhance the system's accuracy and effectiveness.

    Doguhan Ilter

    Thank you for visiting my repository, and I wish you good health!