![]() reported that the AKI predictor (available on the web) predicted AKI with a higher accuracy than that of physicians’ predictions. In risk assessment in the renal field, there were 39 reports on predicting the risk of developing acute kidney injury (AKI) as of March 2021, and Flechet et al. The classification accuracy in the field of image diagnosis is high, and the accuracy has already surpassed that of humans. Supervised learning, which is a subcategory of machine learning, is divided into two types: classification, which predicts discrete values, and regression, which predicts continuous values. Machine learning is a technique for constructing a system to process tasks using big data. Recently, there have been several reports on the application of artificial intelligence (AI) technology in medical care, covering a wide range of areas such as genomic medicine, image diagnosis, diagnostic and therapeutic support, and surgical support. In addition, the method of reciprocal Cr has been reported to have low accuracy. However, these methods require time series data, which is difficult to generate for first-time patients. Conventionally, time series graphs of estimated glomerular filtration rate (eGFR) or reciprocal creatinine (Cr) are used to estimate the time to RRT based on the annual decline rate. In high-risk patients with ESKD, renal function often deteriorates progressively, making it crucial to predict the time to renal replacement therapy (RRT) to achieve a clearer and concrete description of the necessity of treatment. However, CKD patients often delay referral to a nephrologist or discontinue seeing a nephrologist because of a lack of subjective symptoms and their reluctance to continue treatment. The Kidney Disease Improving Global Outcomes (KDIGO) guideline provides a heat map of the risk of progression to end-stage kidney disease (ESKD), and the National Institute for Health and Care Excellence guideline recommends the kidney failure risk equation (KFRE) as a criterion for referral to a nephrologist. Chronic kidney disease (CKD) is a concept proposed for the early detection of renal dysfunction, and it is estimated that 9.1% of the world’s population is affected by CKD. The number of dialysis patients is increasing globally and is expected to be 3.8 million people worldwide by 2021. This approach outperforms the conventional prediction method that uses eGFR time series data and presents new avenues for CKD treatment. ![]() The significance of this study is that it shows that machine learning can predict time to RRT moderately well with continuous values from data at a single time point. ![]() By contrast, the conventional prediction method was found to be extremely low with an R 2 of -17.1. The least absolute shrinkage and selection operator regression model exhibited moderate accuracy with an R 2 of 0.60. Furthermore, we predicted the time to RRT using a conventional prediction method that uses the eGFR decline rate for patients who had measured eGFR three or more times in two years and evaluated its accuracy. We created multiple machine learning models using the training data and evaluated their accuracy using validation data. ![]() The data were preprocessed and split into training and validation datasets. A new machine learning predictor was compared with the established prediction method that uses the eGFR decline rate and the accuracy of the prediction models was determined using the coefficient of determination (R 2). ![]() Methodsĭata of adult chronic kidney disease (CKD) patients who underwent hemodialysis at Oita University Hospital from April 2016 to March 2021 were extracted from electronic medical records ( N = 135). We developed and validated machine learning models for predicting the time to RRT and compared its accuracy with conventional prediction methods that uses the rate of estimated glomerular filtration rate (eGFR) decline. Predicting time to renal replacement therapy (RRT) is important in patients at high risk for end-stage kidney disease. ![]()
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |