Kidney stones are collections of crystals formed from minerals and other substances within the urinary tract. Early identification of kidney stones is important because it may help avoiding complications, enhancing treatment outcomes, and boosting overall quality of life. Identifying kidney stones involves various imaging, laboratory, and diagnostic techniques. Machine learning (ML) and deep learning (DL) techniques have demonstrated their capabilities in identifying kidney stones from radiographic imaging (i.e., CT Scan, X-ray, MRI, etc.). However, ML and DL techniques are yielding good results in not only detecting kidney stone but also their size and locations as well. Kidney stone identification and classification are often addressed in the literature using multi-modal imaging data where different modalities capture different aspects of the disorder. In recent years, various ML and DL driven solutions have been introduced to address this issue. This paper represents an in-depth evaluation of several machine learning and deep learning techniques employed for the identifying kidney stones from radiographic imaging. Following the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines, we performed a systematic collection of research published in the past decade by eminent publishers. The review specifically discusses the clinical application of multi-modal imaging methods, including computed tomography, X-rays, and ultra-sounds, used in kidney stone detection and classification. It also explores the use of these techniques for recognizing other urological diseases. The methodology and performance metrics for each study are critically examined. The aim of this study is to provide insights into the current status of machine learning and deep learning applications for the identification and classification of kidney stones across different imaging modalities.
1 January 2012 in International Journal of Machine Learning and Computing
The investigation for precision in Stock Market Forecasts (SMF) developments has led financial professionals to explore numerous modeling approaches. This paper investigates a new technique aimed at advancing the precision of Support Vector Machine (SMF) by combining the application of Random Forest (RF) methods with Stochastic Differential Equations (SDEs). Market deviations can be enhanced using the Geometric Brownian Motion (GBM) scheme, but this approach has problems because it denotes that market deviations will stay identical. To solve these challenges and develop how the GBM context more precisely corresponds to the unpredictable stock market factors, Random Forest (RF) is used for variable parameter prediction. The change in migration and deviation variables in GBM procedures has been defined using RF approaches in the present study using historical data from the stock market. It examines the performance of the RF-enhanced GBM method by comparing it to static variables in GBM, the Heston Model, and standard GBM while taking unpredictable chances into account. The quantitative metrics, such as the Sharpe Ratio (SR), Cumulative Returns (CR), and Maximum Drawdown (MD), were computed for a total of five stock markets in the research. When compared to the other models, the RF-enhanced GBM generally displayed a competitive edge in terms of risk-adjusted income and was highly accurate in terms of coming up with precise projections.
1 January 2023 in Journal of Discrete Mathematical Sciences and Cryptography
The biggest challenge in the field of image processing is to recognize documents both in printed and handwritten format. Optical Character Recognition OCR is a type of document image analysis where scanned digital image that contains either machine printed or handwritten script input into an OCR software engine and translating it into an editable machine readable digital text format. A Neural network is designed to model the way in which the brain performs a particular task or function of interest: The neural network is simulated in software on a digital computer. Character Recognition refers to the process of converting printed Text documents into translated Unicode Text. The printed documents available in the form of books, papers, magazines, etc. are scanned using standard scanners which produce an image of the scanned document. Lines are identifying by an algorithm where we identify top and bottom of line. Then in each line character boundaries are calculated by an algorithm then using these calculation, characters is isolated from the image and then we classify each character by basic back propagation. Each image character is comprised of 30*20 pixels. We have used the Back propagation Neural Network for efficient recognition where the errors were corrected through back propagation and rectified neuron values were transmitted by feed-forward method in the neural network of multiple layers.