Real-time prediction of intradialytic relative blood volume: a proof-of-concept for integrated cloud computing infrastructure

Abstract

Background Inadequate refilling from extravascular compartments during hemodialysis can lead to intradialytic symptoms, such as hypotension, nausea, vomiting, and cramping/myalgia. Relative blood volume (RBV) plays an important role in adapting the ultrafiltration rate which in turn has a positive effect on intradialytic symptoms. It has been clinically challenging to identify changes RBV in real time to proactively intervene and reduce potential negative consequences of volume depletion. Leveraging advanced technologies to process large volumes of dialysis and machine data in real time and developing prediction models using machine learning (ML) is critical in identifying these signals. Method We conducted a proof-of-concept analysis to retrospectively assess near real-time dialysis treatment data from in-center patients in six clinics using Optical Sensing Device (OSD), during December 2018 to August 2019. The goal of this analysis was to use real-time OSD data to predict if a patient’s relative blood volume (RBV) decreases at a rate of at least − 6.5 % per hour within the next 15 min during a dialysis treatment, based on 10-second windows of data in the previous 15 min. A dashboard application was constructed to demonstrate how reporting structures may be developed to alert clinicians in real time of at-risk cases. Data was derived from three sources: (1) OSDs, (2) hemodialysis machines, and (3) patient electronic health records. Results Treatment data from 616 in-center dialysis patients in the six clinics was curated into a big data store and fed into a Machine Learning (ML) model developed and deployed within the cloud. The threshold for classifying observations as positive or negative was set at 0.08. Precision for the model at this threshold was 0.33 and recall was 0.94. The area under the receiver operating curve (AUROC) for the ML model was 0.89 using test data. Conclusions The findings from our proof-of concept analysis demonstrate the design of a cloud-based framework that can be used for making real-time predictions of events during dialysis treatments. Making real-time predictions has the potential to assist clinicians at the point of care during hemodialysis.

Artificial intelligence and digital health for volume maintenance in hemodialysis patients

“…Candelieri et al explored the use of decision trees and support vector models in the prediction of heart failure decompensation, finding that a "hyper-solution" framework encompassing an optimized ensemble of support vector models showed an accuracy of 87.35% and sensitivity of 90.91% in the early detection of heart failure decompensation. 60,61 There has been significant progress in predicting complex volume-related events during dialysis using machine learning 56,64,65 (Table 3). Machine learning has been applied to the assessment of dry weight in dialysis patients using small dialysis datasets, [66][67][68][69][70] as well in the prediction of volume-related adverse events 56,64,65 (Table 3).…”

“…60,61 There has been significant progress in predicting complex volume-related events during dialysis using machine learning 56,64,65 (Table 3). Machine learning has been applied to the assessment of dry weight in dialysis patients using small dialysis datasets, [66][67][68][69][70] as well in the prediction of volume-related adverse events 56,64,65 (Table 3). Barbieri et al created a prototype algorithm for the prediction of intradialytic hypotension (IDH) using an artificial neural network consisting of 60 variables to create a multiple endpoint model predicting sessionspecific Kt/V, fluid volume removal, HR, and blood pressure (BP).…”

“…There has been significant progress in predicting complex volume‐related events during dialysis using machine learning 56 , 64 , 65 (Table 3 ). Machine learning has been applied to the assessment of dry weight in dialysis patients using small dialysis datasets, 66 , 67 , 68 , 69 , 70 as well in the prediction of volume‐related adverse events 56 , 64 , 65 (Table 3 ).…”

Digitale Anwendungen in verschiedenen Versorgungsstrukturen der Dialyse

“…Erste Ansätze zur Analyse in Echtzeit während der Dialysebehandlung werden zu den Veränderungen des relativen Blutvolumens und zur Prävention von einer intradialytischen Hypotension verfolgt. Als Datenquellen für eine noch nicht publizierte Studie herangezogen wurden über ein Jahr gesammelte Messparameter des Hämatokritsensors für das relative Blutvolumen, technische Dialysemaschinendaten und die elektronische Patientenakte 8 .…”

High Ultrafiltration Rates and Mortality in Hemodialysis Patients: Current Evidence and Future Steps

“…For example, it is known that high relative blood volume decreases are associated with more intradialytic hypotension and increased mortality (16,17). Feasibility of intradialytic monitoring to predict higher decrease in relative blood volume in the subsequent 15 minutes of HD via the use of optical sensing devices (which give information about hematocrit, oxygen saturation, and intravascular blood volume in real time) and patient data has previously been demonstrated (18). Biofeedback mechanisms using such techniques could be used to guide UFR prescriptions dynamically throughout the course of an HD session.…”