Project Overview

This project focuses on predicting kidney disease using clinical and laboratory data. It demonstrates a complete end-to-end machine learning pipeline, including data preprocessing, exploratory data analysis (EDA), model training, and evaluation.

The goal is to build a baseline classification model that can assist in early risk identification of kidney disease.

Dataset

• File: kidney_disease_dataset.csv
• Records: 20,538
• Features: Clinical & laboratory measurements
• Target: Kidney disease risk category

Features Included

• Age
• Blood Pressure
• Urine Albumin
• Urine Sugar
• Red Blood Cells
• Blood Urea
• White Blood Cell Count
• Red Blood Cell Count
• Hemoglobin
• Target (Disease Risk)

Workflow

1. Data Loading
2. Data Cleaning & Preprocessing
3. Exploratory Data Analysis (EDA)
4. Feature Encoding & Scaling
5. Baseline Model Training (Random Forest)
6. Model Evaluation
7. Feature Importance Analysis

Model Used

• Algorithm: Random Forest Classifier
• Evaluation Metrics: Accuracy, Precision, Recall, F1-score
• Visualization: Confusion Matrix, Feature Importance

Results

The baseline model demonstrates strong predictive performance and highlights important clinical features contributing to kidney disease prediction.

Future Improvements

• Hyperparameter tuning
• Advanced models (XGBoost, Neural Networks)
• Explainable AI (SHAP, LIME)
• Deployment using FastAPI

Technologies Used

• Python
• Pandas, NumPy
• Scikit-learn
• Matplotlib, Seaborn

Author

Danish Zulfiqar
AI & ML Engineer