A Block Chain‐Enabled Novel Intelligent System Analysis for Medical Image Processing of Kidney Stone Prediction Using Deep Learning Techniques and Augmented Reality

Globally, kidney stone disease represents a grave condition which impacts millions of individuals. When stone problems are not identified in their infancy, damage to the kidneys develops. Most people in the present situation have kidney failure as a consequence of glomerulonephritis, diabetes, hypertension, more sleeping drugs, and other conditions. This is crucial to receive an evaluation as rapidly as practicable for renal disease due to its potential for harm. Excess calcium salts and uric acid secretion cause kidney stones to develop. The urinary tract may get blocked, resulting in acute pain in the lower back or abdomen. As a result, it is critical to locate the stone to avoid further health problems. The major aim of the proposed work is employing several deep learning algorithms for predicting the presence of kidney stones from a dataset that contains the image of the kidney with low contrast and speckle noise, making sure of the exact volume and position of kidney stones. The proposed system proposes three machine learning models to predict the presence of kidney stones as a hybrid model to get better accuracy results. Blockchain enabled a system for storing the records of patients about the drugs used for treating kidney stones. This technique increases the viability and usability along with the algorithms used in the hybrid model including K-means clustering, support vector machine (SVM), and convolution neural network with long short-term memory (CNN-LSTM). The accuracy received by processing these algorithms is 90%, 93%, and 97% respectively. The CNN-LSTM algorithm is considered an efficient algorithm with accuracy that is higher than the other algorithms. The results obtained through the model is fed into an augmented reality module to enhance the interactivity among users.