Partially Lazy Classification of Cardiovascular Risk via Multi-way Graph Cut Optimization

Karma M. Fathalla, Anikó Ekárt, Doina Gherghel

Research output: Contribution to journalConference article

Abstract

Cardiovascular disease (CVD) is considered a leading cause of human mortality with rising trends worldwide. Therefore, early identification of seemingly healthy subjects at risk is a priority. For this purpose, we propose a novel classification algorithm that provides a sound individual risk prediction, based on a non-invasive assessment of retinal vascular function. so-called lazy classification methods offer reduced time complexity by saving model construction time and better adapting to newly available instances, when compared to well-known eager methodS. Lazy methods are widely used due to their simplicity and competitive performance. However, traditional lazy approaches are more vulnerable to noise and outliers, due to their full reliance on the instances' local neighbourhood for classification. In this work, a learning method based on Graph Cut Optimization called GCO mine is proposed, which considers both the local arrangements and the global structure of the data, resulting in improved performance relative to traditional lazy methodS. We compare GCO mine coupled with genetic algorithms (hGCO mine) with established lazy and eager algorithms to predict cardiovascular risk based on Retinal Vessel Analysis (RVA) data. The highest accuracy of 99.52% is achieved by hGCO mine. The performance of GCO mine is additionally demonstrated on 12 benchmark medical datasets from the UCI repository. In 8 out of 12 datasets, GCO mine outperforms its counterpartS. GCO mine is recommended for studies where new instances are expected to be acquired over time, as it saves model creation time and allows for better generalization compared to state of the art methodS.

Original languageEnglish
Pages (from-to)576-585
Number of pages10
JournalProcedia Computer Science
Volume126
DOIs
Publication statusPublished - 28 Aug 2018
Event22nd International Conference on Knowledge-Based and Intelligent Information and Engineering Systems, KES 2018 - Belgrade, Serbia
Duration: 3 Sep 20185 Sep 2018

Fingerprint

Genetic algorithms
Acoustic waves

Bibliographical note

This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0/)

Keywords

  • Cardiovascular disease
  • genetic algorithm
  • graph cut optimization
  • lazy classification
  • Retinal Vessel Analysis

Cite this

Fathalla, Karma M. ; Ekárt, Anikó ; Gherghel, Doina. / Partially Lazy Classification of Cardiovascular Risk via Multi-way Graph Cut Optimization. In: Procedia Computer Science. 2018 ; Vol. 126. pp. 576-585.
@article{0b7b6722f7374b489631959540d4771d,
title = "Partially Lazy Classification of Cardiovascular Risk via Multi-way Graph Cut Optimization",
abstract = "Cardiovascular disease (CVD) is considered a leading cause of human mortality with rising trends worldwide. Therefore, early identification of seemingly healthy subjects at risk is a priority. For this purpose, we propose a novel classification algorithm that provides a sound individual risk prediction, based on a non-invasive assessment of retinal vascular function. so-called lazy classification methods offer reduced time complexity by saving model construction time and better adapting to newly available instances, when compared to well-known eager methodS. Lazy methods are widely used due to their simplicity and competitive performance. However, traditional lazy approaches are more vulnerable to noise and outliers, due to their full reliance on the instances' local neighbourhood for classification. In this work, a learning method based on Graph Cut Optimization called GCO mine is proposed, which considers both the local arrangements and the global structure of the data, resulting in improved performance relative to traditional lazy methodS. We compare GCO mine coupled with genetic algorithms (hGCO mine) with established lazy and eager algorithms to predict cardiovascular risk based on Retinal Vessel Analysis (RVA) data. The highest accuracy of 99.52{\%} is achieved by hGCO mine. The performance of GCO mine is additionally demonstrated on 12 benchmark medical datasets from the UCI repository. In 8 out of 12 datasets, GCO mine outperforms its counterpartS. GCO mine is recommended for studies where new instances are expected to be acquired over time, as it saves model creation time and allows for better generalization compared to state of the art methodS.",
keywords = "Cardiovascular disease, genetic algorithm, graph cut optimization, lazy classification, Retinal Vessel Analysis",
author = "Fathalla, {Karma M.} and Anik{\'o} Ek{\'a}rt and Doina Gherghel",
note = "This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0/)",
year = "2018",
month = "8",
day = "28",
doi = "10.1016/j.procS.2018.07.292",
language = "English",
volume = "126",
pages = "576--585",

}

Partially Lazy Classification of Cardiovascular Risk via Multi-way Graph Cut Optimization. / Fathalla, Karma M.; Ekárt, Anikó; Gherghel, Doina.

In: Procedia Computer Science, Vol. 126, 28.08.2018, p. 576-585.

Research output: Contribution to journalConference article

TY - JOUR

T1 - Partially Lazy Classification of Cardiovascular Risk via Multi-way Graph Cut Optimization

AU - Fathalla, Karma M.

AU - Ekárt, Anikó

AU - Gherghel, Doina

N1 - This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0/)

PY - 2018/8/28

Y1 - 2018/8/28

N2 - Cardiovascular disease (CVD) is considered a leading cause of human mortality with rising trends worldwide. Therefore, early identification of seemingly healthy subjects at risk is a priority. For this purpose, we propose a novel classification algorithm that provides a sound individual risk prediction, based on a non-invasive assessment of retinal vascular function. so-called lazy classification methods offer reduced time complexity by saving model construction time and better adapting to newly available instances, when compared to well-known eager methodS. Lazy methods are widely used due to their simplicity and competitive performance. However, traditional lazy approaches are more vulnerable to noise and outliers, due to their full reliance on the instances' local neighbourhood for classification. In this work, a learning method based on Graph Cut Optimization called GCO mine is proposed, which considers both the local arrangements and the global structure of the data, resulting in improved performance relative to traditional lazy methodS. We compare GCO mine coupled with genetic algorithms (hGCO mine) with established lazy and eager algorithms to predict cardiovascular risk based on Retinal Vessel Analysis (RVA) data. The highest accuracy of 99.52% is achieved by hGCO mine. The performance of GCO mine is additionally demonstrated on 12 benchmark medical datasets from the UCI repository. In 8 out of 12 datasets, GCO mine outperforms its counterpartS. GCO mine is recommended for studies where new instances are expected to be acquired over time, as it saves model creation time and allows for better generalization compared to state of the art methodS.

AB - Cardiovascular disease (CVD) is considered a leading cause of human mortality with rising trends worldwide. Therefore, early identification of seemingly healthy subjects at risk is a priority. For this purpose, we propose a novel classification algorithm that provides a sound individual risk prediction, based on a non-invasive assessment of retinal vascular function. so-called lazy classification methods offer reduced time complexity by saving model construction time and better adapting to newly available instances, when compared to well-known eager methodS. Lazy methods are widely used due to their simplicity and competitive performance. However, traditional lazy approaches are more vulnerable to noise and outliers, due to their full reliance on the instances' local neighbourhood for classification. In this work, a learning method based on Graph Cut Optimization called GCO mine is proposed, which considers both the local arrangements and the global structure of the data, resulting in improved performance relative to traditional lazy methodS. We compare GCO mine coupled with genetic algorithms (hGCO mine) with established lazy and eager algorithms to predict cardiovascular risk based on Retinal Vessel Analysis (RVA) data. The highest accuracy of 99.52% is achieved by hGCO mine. The performance of GCO mine is additionally demonstrated on 12 benchmark medical datasets from the UCI repository. In 8 out of 12 datasets, GCO mine outperforms its counterpartS. GCO mine is recommended for studies where new instances are expected to be acquired over time, as it saves model creation time and allows for better generalization compared to state of the art methodS.

KW - Cardiovascular disease

KW - genetic algorithm

KW - graph cut optimization

KW - lazy classification

KW - Retinal Vessel Analysis

UR - http://www.scopus.com/inward/record.url?scp=85056656411&partnerID=8YFLogxK

UR - https://www.sciencedirect.com/science/article/pii/S1877050918312687?via%3Dihub

U2 - 10.1016/j.procS.2018.07.292

DO - 10.1016/j.procS.2018.07.292

M3 - Conference article

AN - SCOPUS:85056656411

VL - 126

SP - 576

EP - 585

ER -