Learning-Based Screening of Endothelial Dysfunction From Photoplethysmographic Signals
Endothelial-Dysfunction (ED) screening is of primary importance to early diagnosis cardiovascular diseases. Recently, approaches to ED screening are focusing more and more on photoplethysmography (PPG)-signal analysis, which is performed in a threshold-sensitive way and may not be suitable for tackl...
Main Authors: | , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2019-03-01
|
Series: | Electronics |
Subjects: | |
Online Access: | http://www.mdpi.com/2079-9292/8/3/271 |
id |
doaj-0d6f8ddaf4de44fe86dd8779c7ac7d4b |
---|---|
record_format |
Article |
spelling |
doaj-0d6f8ddaf4de44fe86dd8779c7ac7d4b2020-11-25T00:30:03ZengMDPI AGElectronics2079-92922019-03-018327110.3390/electronics8030271electronics8030271Learning-Based Screening of Endothelial Dysfunction From Photoplethysmographic SignalsChiara Calamanti0Sara Moccia1Lucia Migliorelli2Marina Paolanti3Emanuele Frontoni4Department of Information Engineering, Universitá Politecnica delle Marche, 60121 Ancona, ItalyDepartment of Information Engineering, Universitá Politecnica delle Marche, 60121 Ancona, ItalyDepartment of Information Engineering, Universitá Politecnica delle Marche, 60121 Ancona, ItalyDepartment of Information Engineering, Universitá Politecnica delle Marche, 60121 Ancona, ItalyDepartment of Information Engineering, Universitá Politecnica delle Marche, 60121 Ancona, ItalyEndothelial-Dysfunction (ED) screening is of primary importance to early diagnosis cardiovascular diseases. Recently, approaches to ED screening are focusing more and more on photoplethysmography (PPG)-signal analysis, which is performed in a threshold-sensitive way and may not be suitable for tackling the high variability of PPG signals. The goal of this work was to present an innovative machine-learning (ML) approach to ED screening that could tackle such variability. Two research hypotheses guided this work: (H1) ML can support ED screening by classifying PPG features; and (H2) classification performance can be improved when including also anthropometric features. To investigate H1 and H2, a new dataset was built from 59 subject. The dataset is balanced in terms of subjects with and without ED. Support vector machine (SVM), random forest (RF) and k-nearest neighbors (KNN) classifiers were investigated for feature classification. With the leave-one-out evaluation protocol, the best classification results for H1 were obtained with SVM (accuracy = 71%, recall = 59%). When testing H2, the recall was further improved to 67%. Such results are a promising step for developing a novel and intelligent PPG device to assist clinicians in performing large scale and low cost ED screening.http://www.mdpi.com/2079-9292/8/3/271endothelial dysfunctionphotoplethysmographymachine learningcomputer-assisted screening |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Chiara Calamanti Sara Moccia Lucia Migliorelli Marina Paolanti Emanuele Frontoni |
spellingShingle |
Chiara Calamanti Sara Moccia Lucia Migliorelli Marina Paolanti Emanuele Frontoni Learning-Based Screening of Endothelial Dysfunction From Photoplethysmographic Signals Electronics endothelial dysfunction photoplethysmography machine learning computer-assisted screening |
author_facet |
Chiara Calamanti Sara Moccia Lucia Migliorelli Marina Paolanti Emanuele Frontoni |
author_sort |
Chiara Calamanti |
title |
Learning-Based Screening of Endothelial Dysfunction From Photoplethysmographic Signals |
title_short |
Learning-Based Screening of Endothelial Dysfunction From Photoplethysmographic Signals |
title_full |
Learning-Based Screening of Endothelial Dysfunction From Photoplethysmographic Signals |
title_fullStr |
Learning-Based Screening of Endothelial Dysfunction From Photoplethysmographic Signals |
title_full_unstemmed |
Learning-Based Screening of Endothelial Dysfunction From Photoplethysmographic Signals |
title_sort |
learning-based screening of endothelial dysfunction from photoplethysmographic signals |
publisher |
MDPI AG |
series |
Electronics |
issn |
2079-9292 |
publishDate |
2019-03-01 |
description |
Endothelial-Dysfunction (ED) screening is of primary importance to early diagnosis cardiovascular diseases. Recently, approaches to ED screening are focusing more and more on photoplethysmography (PPG)-signal analysis, which is performed in a threshold-sensitive way and may not be suitable for tackling the high variability of PPG signals. The goal of this work was to present an innovative machine-learning (ML) approach to ED screening that could tackle such variability. Two research hypotheses guided this work: (H1) ML can support ED screening by classifying PPG features; and (H2) classification performance can be improved when including also anthropometric features. To investigate H1 and H2, a new dataset was built from 59 subject. The dataset is balanced in terms of subjects with and without ED. Support vector machine (SVM), random forest (RF) and k-nearest neighbors (KNN) classifiers were investigated for feature classification. With the leave-one-out evaluation protocol, the best classification results for H1 were obtained with SVM (accuracy = 71%, recall = 59%). When testing H2, the recall was further improved to 67%. Such results are a promising step for developing a novel and intelligent PPG device to assist clinicians in performing large scale and low cost ED screening. |
topic |
endothelial dysfunction photoplethysmography machine learning computer-assisted screening |
url |
http://www.mdpi.com/2079-9292/8/3/271 |
work_keys_str_mv |
AT chiaracalamanti learningbasedscreeningofendothelialdysfunctionfromphotoplethysmographicsignals AT saramoccia learningbasedscreeningofendothelialdysfunctionfromphotoplethysmographicsignals AT luciamigliorelli learningbasedscreeningofendothelialdysfunctionfromphotoplethysmographicsignals AT marinapaolanti learningbasedscreeningofendothelialdysfunctionfromphotoplethysmographicsignals AT emanuelefrontoni learningbasedscreeningofendothelialdysfunctionfromphotoplethysmographicsignals |
_version_ |
1725328253023944704 |