Artificial intelligence and digital health for volume maintenance in hemodialysis patients

Chronic fluid overload is associated with morbidity and mortality in hemodialysis patients. Optimizing the diagnosis and treatment of fluid overload remains a priority for the nephrology community. Although current methods of assessing fluid status, such as bioimpedance and lung ultrasound, have prognostic and diagnostic value, no single system or technique can be used to maintain euvolemia. The difficulty in maintaining and assessing fluid status led to a publication by the Kidney Health Initiative in 2019 aimed at fostering innovation in fluid management therapies. This review article focuses on the current limitations in our assessment of extracellular volume, and the novel technology and methods that can create a new paradigm for fluid management. The cardiology community has published research on multiparametric wearable devices that can create individualized predictions for heart failure events. In the future, similar wearable technology may be capable of tracking fluid changes during the interdialytic period and enabling behavioral change. Machine learning methods have shown promise in the prediction of volume-related adverse events. Similar methods can be leveraged to create accurate, automated predictions of dry weight that can potentially be used to guide ultrafiltration targets and interdialytic weight gain goals.