Chronic Kidney Disease as a Cardiovascular Disorder-Tonometry Data Analyses

Area of Science:

• Nephrology
• Cardiology
• Biomedical Engineering
• Artificial Intelligence

Background:

• Tonometry is a cost-effective, non-invasive method for cardiovascular disease (CVD) diagnostics.
• Chronic kidney disease (CKD) and end-stage renal disease (ESRD) significantly impact patient health and require accurate diagnostic tools.
• The potential of tonometry data to reflect different stages of CKD and ESRD treatment remains largely unexplored.

Purpose of the Study:

• To investigate if tonometry data contains unique profiles associated with various stages of CKD and ESRD.
• To assess the ability of artificial intelligence (AI) and statistical methods to differentiate between patient groups based on tonometry data.
• To compare the efficacy of different analytical techniques in classifying kidney disease stages.

Main Methods:

• Six patient groups were assessed: varying stages of CKD on different dialysis treatments, CVD patients without CKD, and healthy controls.
• Statistical analyses including analysis of variance, network correlation structure, multinomial logistic regression, and discrimination analysis were employed.
• AI and machine learning approaches were utilized to build and evaluate classification models.

Main Results:

• Each patient group exhibited a distinct tonometric profile.
• The type of dialysis (hemodialysis vs. peritoneal dialysis) was not distinguishable via tonometry.
• AI models demonstrated the capability to differentiate between various stages of CKD and non-CKD patients.

Conclusions:

• Tonometry data, analyzed by AI, holds significant potential for classifying different stages of kidney disease.
• Future machine learning models may accurately determine kidney health and classify patients, aiding clinical decision-making.
• Further research is needed to translate these findings into simple mathematical relations for clinical application.