Downloads: 1 | Views: 122 | Weekly Hits: ⮙1 | Monthly Hits: ⮙1

Research Paper | Computer Science & Engineering | India | Volume 12 Issue 6, June 2023

A Machine Learning Approach for the Diagnosis of Chronic Kidney Disease

Divya Pogaku | Sneha Bohra ^[2]

Abstract: Several billion people worldwide are afflicted by chronic kidney disease (CKD), a common and possibly fatal ailment. For efficient management and treatment of CKD, an early and precise diagnosis is essential. Machine learning (ML) algorithms have recently demonstrated promising results in a number of medical fields. This work uses a big dataset of clinical, demographic, and laboratory data gathered from a large cohort of CKD patients to provide a machine learning method to CKD diagnosis. To ensure the dependability and robustness of the ML models, the dataset handle values that are missing, outliers, and class imbalances during pre - processing. Numerous machine learning techniques encompass k - nearest neighbors (KNN), Random Forest, logistic regression, support vector machines (SVM), Naive Bayes classifiers, and feed - forward neural networks. Cross - validation technique is applied to train and test feed forward neural networks. Model performance is evaluated by estimating the accuracy of the model, with the Random Forest machine learning model achieving the highest accuracy. With the use of perceptron, we suggested a combined model that combines random forest with logistic regression which has the best accuracy; as a result, we hypothesized that more complicated clinical data may be used with this technology to diagnose disorders.

Keywords: KNN, Naive Bayes classifier, Logistic Regression, Random Forest and SVM

Edition: Volume 12 Issue 6, June 2023,

Pages: 1168 - 1174