A multi-layer monitoring system for clinical management of Congestive Heart Failure

Abstract

BackgroundCongestive Heart Failure (CHF) is a serious cardiac condition that brings high risks of urgent hospitalization and death. Remote monitoring systems are well-suited to managing patients suffering from CHF, and can reduce deaths and re-hospitalizations, as shown by the literature, including multiple systematic reviews.MethodsThe monitoring system proposed in this paper aims at helping CHF stakeholders make appropriate decisions in managing the disease and preventing cardiac events, such as decompensation, which can lead to hospitalization or death. Monitoring activities are stratified into three layers: scheduled visits to a hospital following up on a cardiac event, home monitoring visits by nurses, and patient's self-monitoring performed at home using specialized equipment. Appropriate hardware, desktop and mobile software applications were developed to enable a patient's monitoring by all stakeholders. For the first two layers, we designed and implemented a Decision Support System (DSS) using machine learning (Random Forest algorithm) to predict the number of decompensations per year and to assess the heart failure severity based on a variety of clinical data. For the third layer, custom-designed sensors (the Blue Scale system) for electrocardiogram (EKG), pulse transit times, bio-impedance and weight allowed frequent collection of CHF-related data in the comfort of the patient's home.We also performed a short-term Heart Rate Variability (HRV) analysis on electrocardiograms self-acquired by 15 healthy volunteers and compared the obtained parameters with those of 15 CHF patients from PhysioNet's PhysioBank archives.ResultsWe report numerical performances of the DSS, calculated as multiclass accuracy, sensitivity and specificity in a 10-fold cross-validation. The obtained average accuracies are: 71.9% in predicting the number of decompensations and 81.3% in severity assessment. The most serious class in severity assessment is detected with good sensitivity and specificity (0.87 / 0.95), while, in predicting decompensation, high specificity combined with good sensitivity prevents false alarms. The HRV parameters extracted from the self-measured EKG using the Blue Scale system of sensors are comparable with those reported in the literature about healthy people.ConclusionsThe performance of DSSs trained with new patients confirmed the results of previous work, and emphasizes the strong correlation between some CHF markers, such as brain natriuretic peptide (BNP) and ejection fraction (EF), with the outputs of interest. Comparing HRV parameters from healthy volunteers with HRV parameters obtained from PhysioBank archives, we confirm the literature that considers the HRV a promising method for distinguishing healthy from CHF patients.

Heart Failure: Diagnosis, Severity Estimation and Prediction of Adverse Events Through Machine Learning Techniques

“…Recently the authors of [33] proposed a multi-layer monitoring system for clinical management of CHF. The three layers include the following monitor activities: a) scheduled visits to a hospital following up the presence of a HF event, b) home monitoring visits by nurses, and c) patient's self-monitoring at home through the utilization of specialized equipment.…”

“…The prediction of the destabilization of HF patients was also addressed by Guidi et al 2014 [32] and Guidi et al 2015 [33]. They made a prediction of the frequency (none, rare or frequent) of CHF decompensation during the year after the first visit using five machine learning techniques (NN, SVM, Fuzzy -Genetic Expert System, Random Forests and CART).…”

“…However, in terms of critical error the best results were produced by the Random Forest algorithm. In Guidi et al 2015 [33], the prediction was addressed as three different classification problems, none vs. all, rare vs. all and frequent vs. all, employing the Random Forests algorithm. The overall accuracy produced by the 10-fold cross-validation procedure is 71.90%, while the sensitivity and specificity for each case that was studied is 57% and 79% for the first case, 65% and 60% for the second case and 59% and 96% for the third case.…”

Artificial intelligence in medical devices and clinical decision support systems

“…Similarly, patients with CHF can be given an electronic "safety net" to enable safer, more affordable, and potentially better quality of life using remote AI/ML monitoring systems (Guidi, 2015;Guidi, 2016). Also, in 2017, the first prescription-only app was approved by the US FDA (Software as a Medical Device) to support patients with substance abuse management (Waltz, 2017).…”

Estimation of New York Heart Association class in heart failure patients based on machine learning techniques

“…For the evaluation of the classifiers, a subject based cross validation approach was followed to address the fact that the dataset included correlated data (baseline and follow-up data of the same patient). In [8] a multi-layer monitoring system for clinical management of congestive HF (CHF) is presented where a decision support system was developed providing prediction of de-compensations and assessment of the HF severity based on the RF algorithm. A scoring model allowing classification of a subject to three groups, healthy group (without cardiac dysfunction), HF-prone group (asymptomatic stages of cardiac dysfunction) and HF group (symptomatic stages of cardiac dysfunction) was presented by Yang et al [9].…”