❤️ Heart Disease Diagnosis using Machine Learning and Data Mining

📋 Project Overview

This repository contains the complete implementation of my MSc dissertation project: "Heart Disease Diagnosis using Machine Learning and Data Mining." The goal is to develop predictive models that accurately diagnose heart disease from patient clinical data, helping to enhance early detection and healthcare outcomes.

The project tackles challenges such as imbalanced data and feature relevance to build robust, interpretable machine learning models that can assist clinicians in decision-making.

🎯 Objectives

Cleanse, normalize, and select the most impactful features from clinical datasets.
Build and compare multiple supervised classification models:
- Random Forest
- Decision Tree
- Naive Bayes
- Support Vector Machine (SVM)
Evaluate model performance using comprehensive metrics:
Accuracy, Precision, Recall, F1-score, Confusion Matrix, and ROC-AUC.
Identify key clinical features driving heart disease prediction.
Develop a simple Tkinter-based GUI to enable easy, interactive patient risk prediction.

📊 Dataset

The dataset is a combined collection sourced from IEEEDataport incorporating records from:

Cleveland Clinic Foundation
Hungarian Institute of Cardiology
University Hospital, Zurich
V.A. Medical Center, Long Beach
Statlog (Heart) Dataset

🔍 Features

The dataset contains 11 clinical features, including:

Age
Sex
Chest Pain Type
Resting Blood Pressure
Cholesterol
Fasting Blood Sugar
Resting ECG Results
Maximum Heart Rate
Exercise-Induced Angina
Oldpeak (ST Depression)
ST Slope

⚙️ Methodology

Data Preprocessing

Handle missing values
Normalize and scale features
Select relevant features using statistical tests and model-based rankings

Model Development

Train Decision Tree, Random Forest, Naive Bayes, and SVM classifiers

Model Evaluation

Assess performance with Accuracy, Precision, Recall, F1-score, Confusion Matrix, and ROC-AUC curves

Feature Importance

Analyze and aggregate feature importance across models to identify critical predictors

🖥️ GUI Application

Built with Tkinter
User-friendly interface for inputting patient data
Provides heart disease prediction based on the trained Random Forest model

🛠️ Technologies Used

🌟 About Me

I’m Daniel Toluwani Adeleke, a Data Scientist & IT professional with a passion for building end-to-end data solutions.
I hold a BSc in Computer Science and an MSc in Data Science & Business Analytics. My expertise includes SQL, Python, Machine Learning, and BI reporting.

📧 Email: dannydave1000@gmail.com
💼 LinkedIn: linkedin.com/in/dannydave
🌐 Portfolio: dannydave.my_portfolio.github.io

dannydave/Heart-Disease-Diagnosis-using-Machine-Learning-and-Data-Mining