Early prediction of nephropathy chronic with supervised machine learning algorithms and feature selection method | IEEE Conference Publication | IEEE Xplore

Early prediction of nephropathy chronic with supervised machine learning algorithms and feature selection method


Abstract:

The application of technology for detecting chronic kidney disease was vital in preventing serious health complications. In today’s medical landscape, this technological ...Show More

Abstract:

The application of technology for detecting chronic kidney disease was vital in preventing serious health complications. In today’s medical landscape, this technological advancement empowered physicians to promptly and accurately interpret diseases, thereby improving patient care and outcomes. Predicting chronic kidney diseases held immense importance due to its significant implications for public health. Despite the efforts of numerous researchers over the years to develop precise prediction models, this field encountered significant challenges stemming from inadequate relevant datasets and suitable prediction methodologies. This research endeavored to tackle these challenges by concentrating on predictive analysis within the healthcare sector, with specific focus on chronic kidney diseases. Two models of supervised machine learning, K-Nearest Neighbors and Gradient Boosting, were employed, utilizing the feature selection method LASSO. The performance of these models was assessed using a separate test dataset. According to our findings, the supervised Gradient Boosting model demonstrated the highest accuracy, indicating its effectiveness in predicting CKD. This underscored the potential of advanced machine learning techniques in improving early detection and management of chronic nephropathy, thereby contributing to better healthcare outcomes.
Date of Conference: 24-25 April 2024
Date Added to IEEE Xplore: 03 June 2024
ISBN Information:
Conference Location: EL OUED, Algeria

References

References is not available for this document.