Congestive Heart Failure (CHF) is a serious cardiac condition that brings high risks of urgent hospitalization and death. Remote monitoring systems are well-suited to manage patients suffering from CHF and can reduce deaths and re-hospitalizations, as shown by the literature, including multiple systematic reviews. The monitoring system proposed in this thesis aims to help CHF stakeholders to make appropriate decisions in managing the disease and preventing cardiac events, such as decompensation, which can lead to hospitalization or death. The whole thesis work is composed of a part of research about the analysis of the typical CHF clinical pathways and its monitoring procedures, and a part of research and innovative development that aims to create software and models (machine learning) useful to the various stakeholders of care processes. In order to include our system in a feasible clinical pathway we proposed a CHF monitoring scenario stratified into three layers: 1-Hospital scheduled visits performed by cardiologist, 2-home monitoring visits performed by nurses, and 3-home monitoring measurements performed by the patient using specialized equipment. Appropriate desktop and mobile software applications were developed in this thesis work to enable such multilayer CHF monitoring. 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 exacerbations per year and to assess the CHF severity based on a variety of clinical data. This represents the research core of this thesis. Performances of the trained machine learning techniques are established using k-folds Cross Validation method and, if compared with literature, results are good (81.3% multiclass-accuracy in severity assessment and 71.9% in prediction of exacerbations). For the third layer, we contacted the University of Houston who has developed some 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. It was then performed a study on possible commercial cloud “analytics as a service” solutions to process biometric signals and to build a predictor system for the early detection of Heart Failure.

System for aiding clinical management of congestive heart failure to improve patient assistance at home / Guidi, Gabriele. - (2017).

System for aiding clinical management of congestive heart failure to improve patient assistance at home.

GUIDI, GABRIELE
2017

Abstract

Congestive Heart Failure (CHF) is a serious cardiac condition that brings high risks of urgent hospitalization and death. Remote monitoring systems are well-suited to manage patients suffering from CHF and can reduce deaths and re-hospitalizations, as shown by the literature, including multiple systematic reviews. The monitoring system proposed in this thesis aims to help CHF stakeholders to make appropriate decisions in managing the disease and preventing cardiac events, such as decompensation, which can lead to hospitalization or death. The whole thesis work is composed of a part of research about the analysis of the typical CHF clinical pathways and its monitoring procedures, and a part of research and innovative development that aims to create software and models (machine learning) useful to the various stakeholders of care processes. In order to include our system in a feasible clinical pathway we proposed a CHF monitoring scenario stratified into three layers: 1-Hospital scheduled visits performed by cardiologist, 2-home monitoring visits performed by nurses, and 3-home monitoring measurements performed by the patient using specialized equipment. Appropriate desktop and mobile software applications were developed in this thesis work to enable such multilayer CHF monitoring. 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 exacerbations per year and to assess the CHF severity based on a variety of clinical data. This represents the research core of this thesis. Performances of the trained machine learning techniques are established using k-folds Cross Validation method and, if compared with literature, results are good (81.3% multiclass-accuracy in severity assessment and 71.9% in prediction of exacerbations). For the third layer, we contacted the University of Houston who has developed some 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. It was then performed a study on possible commercial cloud “analytics as a service” solutions to process biometric signals and to build a predictor system for the early detection of Heart Failure.
2017
Dino Giuli
ITALIA
Guidi, Gabriele
File in questo prodotto:
File Dimensione Formato  
Guidi_Gabriele Tesi Dottorato XXIX Ciclo.pdf

accesso aperto

Tipologia: Tesi di dottorato
Licenza: Open Access
Dimensione 6.89 MB
Formato Adobe PDF
6.89 MB Adobe PDF

I documenti in FLORE sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificatore per citare o creare un link a questa risorsa: https://hdl.handle.net/2158/1080160
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
social impact