Diagnostic Assessment and Interpretation of Clinical Kidney Dataset Using Machine Learning

Résumé fourni par la source

One of the vast applications of machine learning is in medical science. The number of individuals who face the medical condition of kidney failure is critical. One of the problems is the inadequate interpretation of medical diagnostics. This study employed machine learning to proffer solutions to this challenge. The design of an ML model that can detect, classify, and interpret kidney test data was achieved using different machine learning algorithms. This was carried out by training an ML model using a kidney test dataset and different machine learning algorithms. The different machine learning algorithms applied were Random Forest, Naïve-Bayes, Support Vector Machine, and Kernel Support Vector Machine, to establish the most efficient algorithm and therefore indicate that machine learning can classify and interpret kidney test datasets. Random Forest algorithms have the best test accuracy, which is 100%, Support Vector Machine and Kernel Support Vector Machine algorithms have a close accuracy score close to that of Random Forest, with the test accuracy of 95.83% and train accuracy of 93.2%, both having the same accuracy value. The Naïve-Bayes algorithm has the lowest accuracy score of 93% for both the tested and trained data. This finding established the integrated application of artificial intelligence and, therefore, indicates that machine learning can classify and categorize the chronic kidney disease dataset.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.