Skip to content

Latest commit

 

History

History
278 lines (225 loc) · 8.5 KB

README.md

File metadata and controls

278 lines (225 loc) · 8.5 KB
Raw

Parkinson's Disease Prediction App 🧠 🔍

forthebadge made-with-python

Parkinson's disease (PD) is a neurodegenerative disorder that affects movement control. This project leverages machine learning techniques to predict the likelihood of an individual having Parkinson's disease based on their medical features. The model is trained on a dataset containing medical records, and a Streamlit app provides a user-friendly interface to interact with the prediction model. 🧬✨

Deployed Web Application 🌐

🔗 Parkinson's Disease Prediction

About This App 🖥️

This application uses a machine learning model to predict whether an individual has Parkinson's disease based on various vocal feature inputs. 🎤🔊 The app allows users to interactively input medical data and receive a prediction.

About This Repository 📂

This repository contains:

  • A trained machine learning model for predicting Parkinson's disease.
  • A Streamlit app that provides an interactive interface for predictions.

Project Structure 🛠️

The project is organized into different components, including model training, data processing, and the Streamlit web app. Here’s the directory structure:

├── model_files/           # Folder containing model files and other relevant files
│   ├── parkinson_model.pkl  # Saved trained model in pickle format
│   ├── parkinson_model.sav  # Another format of the trained model
│   ├── pca.pkl             # Principal Component Analysis (PCA) model
│   ├── scaler.pkl         # StandardScaler model used during training
|
├── data/                  # Folder for dataset and related files
│   └── parkinsons.data     # Dataset used for model training
|
├── requirements.txt       # Python dependencies required to run the project
└── oldrequirements1.txt   # Old version of requirements file (if needed)
|
├── app8.py                # Streamlit app for interactive prediction
├── dv_cp_.py              # Helper functions and data preprocessing code
|
| etc...

Machine Learning Models Trained & Evaluated 🧑‍💻

The following Machine Learning models were trained and evaluated:
1️⃣ Logistic Regression
2️⃣ Random Forest Classifier
3️⃣ Decision Tree Classifier
4️⃣ Support Vector Machine Classifier
5️⃣ Naive Bayes Classifier
6️⃣ K Nearest Neighbor Classifier

Parkinson's Disease Data Set Description 📊

Data Set Characteristics Multivariate
Number of Instances 197
Area Life
Attribute Characteristics Real
Number of Attributes 23
Date Donated 2008-06-26
Associated Task Classification
Missing Values? N/A

Attribute Information

Medical Attribute Information 📋

Attribute Meaning
name ASCII subject name and recording number
MDVP:Fo(Hz) Average vocal fundamental frequency
MDVP:Fhi(Hz) Maximum vocal fundamental frequency
MDVP:Flo(Hz) Minimum vocal fundamental frequency
MDVP:Jitter(%) Measure of variation in fundamental frequency
MDVP:Jitter(Abs) Measure of variation in fundamental frequency
MDVP:RAP Measure of variation in fundamental frequency
MDVP:PPQ Measure of variation in fundamental frequency
Jitter:DDP Measure of variation in fundamental frequency
MDVP:Shimmer Measure of variation in amplitude
MDVP:Shimmer(dB) Measure of variation in amplitude
Shimmer:APQ3 Measure of variation in amplitude
Shimmer:APQ5 Measure of variation in amplitude
MDVP:APQ Measure of variation in amplitude
Shimmer:DDA Measure of variation in amplitude
NHR Measure of ratio of noise to tonal components in the voice
HNR Measure of ratio of noise to tonal components in the voice
status(Target variable) Health status of the subject (one) - Parkinson's, (zero) - healthy
RPDE Non-linear dynamical complexity measure
D2 Non-linear dynamical complexity measure
DFA Signal fractal scaling exponent
spread1 Non-linear measure of fundamental frequency variation
spread2 Non-linear measure of fundamental frequency variation
PPE Non-linear measure of fundamental frequency variation

Installation 🛠️

1. Clone the repository

git clone https://github.com/your-username/parkinson-disease-prediction.git
cd parkinson-disease-prediction

2. Set up a virtual environment (optional but recommended)

Using venv:

python -m venv venv
source venv/bin/activate   # On Windows, use `venv\Scripts\activate`

3. Install dependencies

pip install -r requirements.txt

Model Training 🔬

The model has been trained using a dataset of medical features of individuals, which can be found in the file parkinsons.data. The training process involves the following steps:

  1. Preprocessing: Data cleaning, handling missing values, and scaling features.
  2. Model Selection: An appropriate machine learning model is trained on the data.
  3. PCA Transformation: Principal Component Analysis (PCA) is applied to reduce dimensionality.
  4. Model Saving: The final trained model and other relevant artifacts like scaler and PCA are saved as pickle files (.pkl, .sav).

You can use these pre-trained models to predict Parkinson's disease on new data by using the Streamlit app.

Using the Streamlit App 🚀

To interact with the trained model and make predictions, we have built a simple Streamlit web app.

Running the app

  1. Navigate to the project directory.

  2. Run the following command:

    streamlit run app8.py

  3. A web browser will automatically open the app, or you can access it at http://localhost:8501 in your browser.

Features of the Streamlit App

  • Input medical features related to the patient.
  • Predict if the individual is likely to have Parkinson's disease.
  • Visualize the prediction result with features importance.

Files 📁

  • parkinson_model.pkl: The trained machine learning model saved using pickle.
  • parkinson_model.sav: An alternate format of the trained model.
  • pca.pkl: Principal Component Analysis model for dimensionality reduction.
  • scaler.pkl: Scaler model used to normalize input data before feeding it into the model.

Contributing 🤝

We welcome contributions to improve the project! If you'd like to contribute, feel free to:

  1. Fork the repository.
  2. Create a new branch for your changes.
  3. Submit a pull request.

Acknowledgements 🙏

  • Dataset: Parkinson's Disease Dataset
  • Libraries used: scikit-learn, streamlit, pandas, and matplotlib.
  • Special thanks to Streamlit Cloud

Group Details 👩‍💻👨‍💻

This project was developed by a group of 4 students from VIT Pune, under the CSAI-B branch.

Roll Number Official Name
33 Shrey Santosh Rupnavar
37 Salitri Atharva Akhil
60 Tanishq Sudhir Thuse
61 Tripti Prakash Mirani

If you have any questions or suggestions, feel free to open an issue or reach out directly! 😄👋