Data collection and correlation analysis of patient-specific dialytic variables for the development of an AI assistance therapeutic tool

Abstract

The increasing prevalence of chronic kidney disease has led to a rise in the number of patients requiring dialytic therapies. These treatments, although essential for blood purification, are often associated with complications such as hypotension, fluid overload, and patient discomfort that traditional models may not fully capture and predict. In this context, data-driven models based on the use of Artificial Intelligence can offer a solution to develop patient-specific dialysis therapies and predict potential complications. This study represents a preliminary step in developing an AI model, focusing on data collection and analysis. Clinical data were collected from 40 patients and statistical analysis, including Pearson's and Spearman's correlation coefficients, has identified significant relationships among several variables, which will be used to implement a model aimed at improving dialysis efficiency, minimizing adverse events, and increasing the overall quality of dialysis patients' care.