Development and nursing application of kidney disease prediction models based on machine learning

Kidney diseases complicate treatment prediction and progression. This study introduces a Metaheuristic Red Fox-Optimized Agile Support Vector Machine (MRFO-ASVM) for early detection and prognosis of kidney diseases. Nurses' involvement in data collection and analysis enhances model effectiveness. Pre-processing with Min-Max normalization and feature extraction using Principal Component Analysis (PCA) improves data quality. The MRFO-ASVM obtained enhanced parameter performance of the model including high accuracy (0.92), F1-score (0.67), sensitivity (0.89), precision (0.63), and ROC-AUC (0.99). Integrating this technology into nursing practice enhances early detection and personalized care, advancing patient-centred healthcare solutions.