Implementation of a Home Monitoring System for Heart Failure Patients: A Feasibility Study

Bioelectrical Impedance Spectroscopy for Monitoring Mammalian Cells and Tissues under Different Frequency Domains: A Review

Bioelectrical impedance analysis and bioelectrical impedance spectroscopy (BIA/BIS) of tissues reveal important information on molecular composition and physical structure that is useful in diagnostics and prognostics. The heterogeneity in structural elements of cells, tissues, organs, and the whole human body, the variability in molecular composition arising from the dynamics of biochemical reactions, and the contributions of inherently electroresponsive components, such as ions, proteins, and polarized membranes, have rendered bioimpedance challenging to interpret but also a powerful evaluation and monitoring technique in biomedicine. BIA/BIS has thus become the basis for a wide range of diagnostic and monitoring systems such as plethysmography and tomography. The use of BIA/BIS arises from (i) being a noninvasive and safe measurement modality, (ii) its ease of miniaturization, and (iii) multiple technological formats for its biomedical implementation. Considering the dependency of the absolute and relative values of impedance on frequency, and the uniqueness of the origins of the α-, β-, δ-, and γ-dispersions, this targeted review discusses biological events and underlying principles that are employed to analyze the impedance data based on the frequency range. The emergence of BIA/BIS in wearable devices and its relevance to the Internet of Medical Things (IoMT) are introduced and discussed.

A Vital Signs Telemonitoring Programme Improves the Dynamic Prediction of Readmission Risk in Patients with Heart Failure

Heart failure (HF) is a leading cause of hospital readmissions. There is great interest in approaches to efficiently predict emerging HF-readmissions in the community setting. We investigate the possibility of leveraging streaming telemonitored vital signs data alongside readily accessible patient profile information for predicting evolving 30-day HF-related readmission risk. We acquired data within a non-randomized controlled study that enrolled 150 HF patients over a 1-year post-discharge telemonitoring and telesupport programme. Using the sequential data and associated ground truth readmission outcomes, we developed a recurrent neural network model for dynamic risk prediction. The model detects emerging readmissions with sensitivity > 71%, specificity > 75%, AUROC ~80%. We characterize model performance in relation to telesupport based nurse assessments, and demonstrate strong sensitivity improvements. Our approach enables early stratification of high-risk patients and could enable adaptive targeting of care resources for managing patients with the most urgent needs at any given time.

Classification of Aortic Stenosis Using Time-Frequency Features From Chest Cardio-Mechanical Signals

OBJECTIVES

This paper introduces a novel method for the detection and classification of aortic stenosis (AS) using the time-frequency features of chest cardio-mechanical signals collected from wearable sensors, namely seismo-cardiogram (SCG) and gyro-cardiogram (GCG) signals. Such a method could potentially monitor high-risk patients out of the clinic.

METHODS

Experimental measurements were collected from twenty patients with AS and twenty healthy subjects. Firstly, a digital signal processing framework is proposed to extract time-frequency features. The features are then selected via the analysis of variance test. Different combinations of features are evaluated using the decision tree, random forest, and artificial neural network methods. Two classification tasks are conducted. The first task is a binary classification between normal subjects and AS patients. The second task is a multi-class classification of AS patients with co-existing valvular heart diseases.

RESULTS

In the binary classification task, the average accuracies achieved are 96.25% from decision tree, 97.43% from random forest, and 95.56% from neural network. The best performance is from combined SCG and GCG features with random forest classifier. In the multi-class classification, the best performance is 92.99% using the random forest classifier and SCG features.

CONCLUSION

The results suggest that the solution could be a feasible method for classifying aortic stenosis, both in the binary and multi-class tasks. It also indicates that most of the important time-frequency features are below 11 Hz.

SIGNIFICANCE

The proposed method shows great potential to provide continuous monitoring of valvular heart diseases to prevent patients from sudden critical cardiac situations.

A Binary Classification of Cardiovascular Abnormality Using Time-Frequency Features of Cardio-mechanical Signals

This paper introduces a novel method of binary classification of cardiovascular abnormality using the time-frequency features of cardio-mechanical signals, namely seismocardiography (SCG) and gyrocardiography (GCG) signals. A digital signal processing framework is proposed which utilizes decision tree and support vector machine methods with features generated by continuous wavelet transform. Experimental measurements were collected from twelve patients with cardiovascular diseases as well as twelve healthy subjects to evaluate the proposed method. Results reveal an overall accuracy of more than 94% with the best performance achieved from SVM classifiers with GCG training features. This suggests that the proposed solution could be a promising method for classifying cardiovascular abnormalities.

A Personalized and Interactive Web-Based Health Care Innovation to Advance the Quality of Life and Care of Patients With Heart Failure (ACQUIRE-HF): A Mixed Methods Feasibility Study

Background

Heart failure (HF) is a progressive, debilitating, and complex disease, and due to an increasing incidence and prevalence, it represents a global health and economic problem. Hence, there is an urgent need to evaluate alternative care modalities to current practice to safeguard a high level of care for this growing population.

Objective

Our goal was to examine the feasibility of engaging patients to use patient-centered and personalized tools coupled with a Web-based, shared care and interactive platform in order to empower and enable them to live a better life with their disease.

Methods

We used a mixed methods, single-center, pre-post design. Patients with HF and reduced left ventricular ejection fraction (n=26) were recruited from the outpatient HF clinic at Odense University Hospital (Svendborg Hospital), Denmark, between October 2015 and March 2016. Patients were asked to monitor their health status via the platform using the standardized, disease-specific measure, the Kansas City Cardiomyopathy Questionnaire (KCCQ), and to register their weight. A subset of patients and nursing staff were interviewed after 3-month follow-up about their experiences with the platform.

Results

Overall, patients experienced improvement in patient-reported health status but deterioration in self-care behavior between baseline and 3-month follow-up. The mean score reflecting patient expectations toward use prior to start of the study was lower (16 [SD 5]) than their actual experiences with use of the platform (21 [SD 5]) after 3-month follow-up. Of all patients, 19 completed both a baseline and follow-up KCCQ. A total of 9 experienced deterioration in their health status (range from 3-34 points), while 10 experienced an improvement (range from 1-23 points). The qualitative data indicated that the majority of patients found the registration and monitoring on the platform useful. Both nursing staff and patients indicated that such monitoring could be a useful tool to engage and empower patients, in particular when patients are just diagnosed with HF.

Conclusions

The use of patient tracking and monitoring of health status in HF using a standardized and validated measure seems feasible and may lead to insights that will help educate, empower, and engage patients more in their own disease management, although it is not suitable for all patients. Nursing staff found the patient-centered tool beneficial as a communication tool with patients but were more reticent with respect to using it as a replacement for the personal contact in the outpatient clinic.