🩺 Disease Prediction System using Streamlit

A Machine Learning–based web application that predicts Diabetes, Heart Disease, and Parkinson’s Disease using patient medical parameters. The system integrates multiple trained ML models into a single user-friendly interface and is deployed on Streamlit Cloud for public accessibility.

🚀 Live Application

🔗 Streamlit Deployment https://deployment-of-diagnosis-support-system-on-cloud-app-mygszfsusb.streamlit.app/

📂 Project Repositories Main Project

🔗 https://github.com/myGithub98478/disease-prediction-system-using-stramlit

Deployment Repository

🔗 https://github.com/myGithub98478/Deployment-of-Diagnosis-Support-System-on-Cloud-Streamlit

📖 Project Overview

Early detection of diseases plays a critical role in improving treatment outcomes and reducing healthcare risks. However, many individuals do not have easy access to early diagnostic tools due to limitations in healthcare infrastructure.

This project aims to build a machine learning-based disease prediction system that can assist users in identifying potential health risks by analyzing medical parameters.

The system uses trained machine learning models and provides predictions through a simple and interactive web interface built with Streamlit.

Users can input medical parameters and instantly receive predictions for multiple diseases.

✨ Features

Predict Diabetes

Predict Heart Disease

Predict Parkinson’s Disease

Interactive Streamlit interface

Machine Learning based predictions

Models saved using Pickle

Fast and lightweight predictions

Cloud deployed web application

🛠️ Technologies Used Category Technology Programming Language Python Machine Learning Scikit-learn Data Processing NumPy, Pandas Web Framework Streamlit Model Serialization Pickle Deployment Streamlit Cloud Version Control Git & GitHub 🤖 Machine Learning Models

The project uses supervised machine learning algorithms trained on medical datasets.

1️⃣ Diabetes Prediction

Input parameters include:

Pregnancies

Glucose Level

Blood Pressure

Skin Thickness

Insulin

BMI

Diabetes Pedigree Function

Age

2️⃣ Heart Disease Prediction

Input parameters include:

Age

Sex

Chest Pain Type

Resting Blood Pressure

Cholesterol Level

Fasting Blood Sugar

Resting ECG

Maximum Heart Rate

Exercise Induced Angina

3️⃣ Parkinson’s Disease Prediction

Input parameters include biomedical voice measurements such as:

MDVP Fo

Jitter

Shimmer

NHR

HNR

RPDE

DFA

These attributes help detect vocal impairments associated with Parkinson’s disease.

🏗️ System Architecture Medical Dataset │ ▼ Data Preprocessing │ ▼ Machine Learning Model Training │ ▼ Model Serialization (Pickle) │ ▼ Streamlit Web Application │ ▼ User Inputs Health Parameters │ ▼ Disease Prediction Output 📁 Project Structure disease-prediction-system │ ├── datasets │ ├── diabetes.csv │ ├── heart.csv │ └── parkinsons.csv │ ├── models │ ├── diabetes_model.pkl │ ├── heart_model.pkl │ └── parkinsons_model.pkl │ ├── app.py ├── requirements.txt └── README.md ⚙️ Installation Guide

Follow these steps to run the project locally.

1️⃣ Clone the Repository git clone https://github.com/myGithub98478/disease-prediction-system-using-stramlit.git 2️⃣ Navigate to the Project Folder cd disease-prediction-system-using-stramlit 3️⃣ Install Dependencies pip install -r requirements.txt 4️⃣ Run the Streamlit Application streamlit run app.py

The application will open automatically in your browser.

📊 Model Evaluation

The machine learning models were evaluated using accuracy metrics.

Model Accuracy Diabetes Prediction ~78% Heart Disease Prediction ~85% Parkinson’s Prediction ~87%

(Accuracy may vary depending on dataset and training conditions.)

✅ Advantages

Early disease risk detection

Easy to use web interface

Multiple disease prediction in one system

Accessible from anywhere via cloud

Lightweight and fast prediction system

⚠️ Limitations

Predictions are based on available datasets

Accuracy depends on data quality

Not a replacement for professional medical diagnosis

Currently limited to three diseases

🔮 Future Improvements

Possible enhancements include:

Adding more disease prediction models

Using Deep Learning algorithms

Improving prediction accuracy with larger datasets

Adding user login system

Integration with hospital databases

Deploying on scalable cloud infrastructure

👨‍💻 Author

GitHub Profile 🔗 https://github.com/myGithub98478

📜 License

This project is created for educational and research purposes only.

⚠️ The predictions generated by this system should not replace professional medical advice or clinical diagnosis.