AI-powered chronic kidney disease detection integrating TabNet and LSTM for precision diagnosis

Area of Science:

• Nephrology
• Artificial Intelligence
• Data Science

Background:

• Chronic kidney disease (CKD) poses a significant global health challenge, necessitating early detection for effective management.
• Current diagnostic methods face limitations with complex CKD datasets, often leading to delayed interventions.
• Existing machine learning models require enhancement to accurately process intricate CKD data.

Purpose of the Study:

• To develop an advanced AI-based system for the early detection and diagnosis of chronic kidney disease (CKD).
• To create a system capable of efficiently handling and analyzing complex healthcare data for improved CKD diagnosis.
• To improve upon existing diagnostic models by leveraging sophisticated AI techniques for enhanced accuracy and efficiency.

Main Methods:

• Data collection and preprocessing, including Expectation Maximization for imputation and One-Class SVM for outlier removal.
• Dimensionality reduction using Principal Component Analysis (PCA) followed by feature selection with TabNet.
• Utilizing Long Short-Term Memory (LSTM) networks to capture temporal dependencies in patient data for abnormality detection.

Main Results:

• The AI system achieved high performance metrics: 98.58% accuracy, 98.89% precision, 98.24% recall, and 98.56% F1-score.
• The proposed model demonstrated superior performance compared to existing methods like XGBoost, SVM, and KNN.
• The system effectively minimized false positives and false negatives in CKD detection.

Conclusions:

• The integration of TabNet for feature selection and LSTM for sequential pattern recognition provides a robust framework for early CKD detection.
• This AI-driven approach offers a significant advancement in managing complex healthcare data for timely and accurate diagnosis of CKD.
• The developed system presents a promising solution for improving patient outcomes through early and precise identification of chronic kidney disease.