A Novel Method for Classifying and Detecting Kidney Stone using MRI based on CNN and Attention Model

When training neural networks, the Back Propagation Network approach is frequently employed. It uses data and images to construct an automated kidney stone classification system. Classification of medical MRI kidney imaging and identification of stones typically involves human assessment. Due of the impracticality of dealing with massive data volumes, this technique is incorrect. Operator error is a potential source of noise contamination in Magnetic Resonance (MR) images. This leads to serious problems with image processing and inaccurate classification characteristics. On the other hand, the use AI, NN, and feature extraction to isolate the target area using back propagation network analysis has yielded impressive results. The objective of this endeavor was to detect kidney stones by use of the Back Propagation Network. In order to accomplish this, data image preprocessing acquires a framework, reduces noise, and removes bed-mats. area, volume of area, and stone's surface area are utilized for feature extraction after the momentum phrase and R<inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">proper</inf> term are employed for picture segmentation. All three of these models—CNN, LSTM, and AGCNN-LSTM were utilized to train the projected model. With a 96% accuracy rate, the suggested model surpassed its predecessor.