IMPROVING CONVOLUTIONAL NEURAL NETWORKS USING SPATIAL ATTENTION FOR ACCURATE CLASSIFICATION OF KIDNEY ABNORMALITIES

Abstract

Accurate diagnosis of kidney abnormalities, such as tumors, stones and cysts is essential for the effective treatment of these diseases. Deep Learning has shown great potential in improving the diagnosis of medical images. In this paper, we present an advanced method for the classification of kidney abnormalities by using Convolutional Neural Networks (CNN) integrated with Spatial Attention mechanisms. Our approach focuses on improving the performance of the classification model by identifying and focusing attention on the important areas of the images. Our model is trained and tested on a dataset of renal scan images. It contains different categories of abnormalities. Experimental results show that the integration of Spatial Attention mechanisms in CNN significantly improves the classification performance compared to traditional methods. This approach provides a powerful tool. It can be used by healthcare professionals to efficiently diagnose kidney abnormalities.