Feature selection for chronic kidney disease classification model using convolutional neural network

A complicated and common medical disorder, chronic kidney disease (CKD) has a big impact on public health. Early and accurate classification of CKD is crucial for effective management and treatment and late of treatment may lead to serious problem that may cause death. However, determining an accurate classification of CKD is challenging. Using a deep learning as a method for the classification of the targeted disease. In this research, an approach for classifying CKD using a Convolutional Neural Network (CNN) will be proposed. Experiments will be conducted to study the performance of CNN in the classification of CKD. After collecting the dataset of CKD, the raw data will be pre-processed before the action of deep learning can be taken. Several feature selections will be done to target the important attribute that contribute to the CKD. After that, CNN model architecture design will be implemented to the experiment in getting the result for performance evaluation. To evaluate the performance of the result, employment of various performance metrics, including accuracy, precision, recall, and F1 score are recorded. The successful implementation of the proposed CNN model for clinical data classification in CKD can have significant clinical implications, enabling personalized treatments, disease monitoring, and facilitating the development of targeted therapies. This research helped advance the practise of precision medicine for CKD and enhance patient outcomes.