A Machine Learning Approach for Chronic Heart Failure Diagnosis

Dafni K. Plati; Evanthia E. Tripoliti; Aris Bechlioulis; Aidonis Rammos; Iliada Dimou; Lampros Lakkas; Chris Watson; Ken McDonald; Mark Ledwidge; Rebabonye Pharithi; Joe Gallagher; Lampros K. Michalis; Yorgos Goletsis; Katerina K. Naka; Dimitrios I. Fotiadis

doi:10.3390/diagnostics11101863

A Machine Learning Approach for Chronic Heart Failure Diagnosis

Dafni K. Plati, Evanthia E. Tripoliti, Aris Bechlioulis, Aidonis Rammos, Iliada Dimou, Lampros Lakkas, Chris Watson, Ken McDonald, Mark Ledwidge, Rebabonye Pharithi, Joe Gallagher, Lampros K. Michalis, Yorgos Goletsis, Katerina K. Naka, Dimitrios I. Fotiadis^*

^*Corresponding author for this work

Research output: Contribution to journal › Article › peer-review

91 Downloads (Pure)

Abstract

The aim of this study was to address chronic heart failure (HF) diagnosis with the application of machine learning (ML) approaches. In the present study, we simulated the procedure that is followed in clinical practice, as the models we built are based on various combinations of feature categories, e.g., clinical features, echocardiogram, and laboratory findings. We also investigated the incremental value of each feature type. The total number of subjects utilized was 422. An ML approach is proposed, comprising of feature selection, handling class imbalance, and classification steps. The results for HF diagnosis were quite satisfactory with a high accuracy (91.23%), sensitivity (93.83%), and specificity (89.62%) when features from all categories were utilized. The results remained quite high, even in cases where single feature types were employed.

Original language	English
Article number	1863
Journal	Diagnostics
Volume	11
Issue number	10
DOIs	https://doi.org/10.3390/diagnostics11101863
Publication status	Published - 10 Oct 2021

Keywords

heart failure
machine learning

Access to Document

10.3390/diagnostics11101863Licence: CC BY

A Machine Learning Approach for Chronic Heart Failure Diagnosis
Copyright 2021 the authors. This is an open access article published under a Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution and reproduction in any medium, provided the author and source are cited.
Final published version, 1.03 MBLicence: CC BY

Cite this

Plati, Dafni K. ; Tripoliti, Evanthia E. ; Bechlioulis, Aris ; Rammos, Aidonis ; Dimou, Iliada ; Lakkas, Lampros ; Watson, Chris ; McDonald, Ken ; Ledwidge, Mark ; Pharithi, Rebabonye ; Gallagher, Joe ; Michalis, Lampros K. ; Goletsis, Yorgos ; Naka, Katerina K. ; Fotiadis, Dimitrios I. / A Machine Learning Approach for Chronic Heart Failure Diagnosis. In: Diagnostics. 2021 ; Vol. 11, No. 10.

@article{1196fa46105b41fcb014da15bc2f057d,

title = "A Machine Learning Approach for Chronic Heart Failure Diagnosis",

abstract = "The aim of this study was to address chronic heart failure (HF) diagnosis with the application of machine learning (ML) approaches. In the present study, we simulated the procedure that is followed in clinical practice, as the models we built are based on various combinations of feature categories, e.g., clinical features, echocardiogram, and laboratory findings. We also investigated the incremental value of each feature type. The total number of subjects utilized was 422. An ML approach is proposed, comprising of feature selection, handling class imbalance, and classification steps. The results for HF diagnosis were quite satisfactory with a high accuracy (91.23%), sensitivity (93.83%), and specificity (89.62%) when features from all categories were utilized. The results remained quite high, even in cases where single feature types were employed.",

keywords = "heart failure, machine learning",

author = "Plati, {Dafni K.} and Tripoliti, {Evanthia E.} and Aris Bechlioulis and Aidonis Rammos and Iliada Dimou and Lampros Lakkas and Chris Watson and Ken McDonald and Mark Ledwidge and Rebabonye Pharithi and Joe Gallagher and Michalis, {Lampros K.} and Yorgos Goletsis and Naka, {Katerina K.} and Fotiadis, {Dimitrios I.}",

year = "2021",

month = oct,

day = "10",

doi = "10.3390/diagnostics11101863",

language = "English",

volume = "11",

journal = "Diagnostics",

issn = "2075-4418",

publisher = "MDPI AG",

number = "10",

}

A Machine Learning Approach for Chronic Heart Failure Diagnosis. / Plati, Dafni K.; Tripoliti, Evanthia E.; Bechlioulis, Aris; Rammos, Aidonis; Dimou, Iliada; Lakkas, Lampros; Watson, Chris; McDonald, Ken; Ledwidge, Mark; Pharithi, Rebabonye; Gallagher, Joe; Michalis, Lampros K.; Goletsis, Yorgos; Naka, Katerina K.; Fotiadis, Dimitrios I.

In: Diagnostics, Vol. 11, No. 10, 1863, 10.10.2021.

Research output: Contribution to journal › Article › peer-review

TY - JOUR

T1 - A Machine Learning Approach for Chronic Heart Failure Diagnosis

AU - Plati, Dafni K.

AU - Tripoliti, Evanthia E.

AU - Bechlioulis, Aris

AU - Rammos, Aidonis

AU - Dimou, Iliada

AU - Lakkas, Lampros

AU - Watson, Chris

AU - McDonald, Ken

AU - Ledwidge, Mark

AU - Pharithi, Rebabonye

AU - Gallagher, Joe

AU - Michalis, Lampros K.

AU - Goletsis, Yorgos

AU - Naka, Katerina K.

AU - Fotiadis, Dimitrios I.

PY - 2021/10/10

Y1 - 2021/10/10

N2 - The aim of this study was to address chronic heart failure (HF) diagnosis with the application of machine learning (ML) approaches. In the present study, we simulated the procedure that is followed in clinical practice, as the models we built are based on various combinations of feature categories, e.g., clinical features, echocardiogram, and laboratory findings. We also investigated the incremental value of each feature type. The total number of subjects utilized was 422. An ML approach is proposed, comprising of feature selection, handling class imbalance, and classification steps. The results for HF diagnosis were quite satisfactory with a high accuracy (91.23%), sensitivity (93.83%), and specificity (89.62%) when features from all categories were utilized. The results remained quite high, even in cases where single feature types were employed.

AB - The aim of this study was to address chronic heart failure (HF) diagnosis with the application of machine learning (ML) approaches. In the present study, we simulated the procedure that is followed in clinical practice, as the models we built are based on various combinations of feature categories, e.g., clinical features, echocardiogram, and laboratory findings. We also investigated the incremental value of each feature type. The total number of subjects utilized was 422. An ML approach is proposed, comprising of feature selection, handling class imbalance, and classification steps. The results for HF diagnosis were quite satisfactory with a high accuracy (91.23%), sensitivity (93.83%), and specificity (89.62%) when features from all categories were utilized. The results remained quite high, even in cases where single feature types were employed.

KW - heart failure

KW - machine learning

U2 - 10.3390/diagnostics11101863

DO - 10.3390/diagnostics11101863

M3 - Article

VL - 11

JO - Diagnostics

JF - Diagnostics

SN - 2075-4418

IS - 10

M1 - 1863

ER -

A Machine Learning Approach for Chronic Heart Failure Diagnosis

Abstract

Keywords

Access to Document

Fingerprint

Cite this