Classifying kidney disease using a dense layers deep learning model.

Area of Science:

• Nephrology
• Artificial Intelligence
• Machine Learning

Background:

• Chronic kidney disease (CKD) poses significant health risks, necessitating early diagnosis and effective management.
• Machine learning and deep learning are increasingly explored for kidney disease detection.
• Deep learning models require specific components like hidden layers, activation functions, optimizers, and epochs for accurate disease identification.

Purpose of the Study:

• To develop and evaluate a highly accurate deep learning model for the automatic detection of chronic kidney disease.
• To assess the performance of the proposed model against established metrics and compare it with other existing models.

Main Methods:

• Utilized a publicly accessible dataset containing 24 independent features.
• Developed dense-layered deep neural networks with an optimized architecture.
• Employed key deep learning parameters including hidden layers, activation functions, optimizers, and epochs.

Main Results:

• The proposed deep learning model achieved an outstanding 99% accuracy.
• Performance was validated by high precision, recall, and F1 scores (0.99).
• Robustness and agreement were confirmed through ROC AUC (0.9821) and Matthews Correlation Coefficient (0.9727).

Conclusions:

• The developed deep learning model demonstrates superior accuracy in chronic kidney disease detection compared to other evaluated models.
• The model's high performance metrics indicate its potential as a reliable tool for clinical application.
• Optimized deep neural network architectures are effective for automated kidney disease diagnosis.