PCA-based polling strategy in machine learning framework for coronary artery disease risk assessment in intravascular ultrasound: A link between carotid and coronary grayscale plaque morphology

Tadashi Araki, Nobutaka Ikeda, Devarshi Shukla, Pankaj K. Jain, Narendra D. Londhe, Vimal K. Shrivastava, Sumit K. Banchhor, Luca Saba, Andrew Nicolaides, Shoaib Shafique, John R. Laird, Jasjit S. Suri

Research output: Contribution to journalArticle

18 Citations (Scopus)

Abstract

Background and objective: Percutaneous coronary interventional procedures need advance planning prior to stenting or an endarterectomy. Cardiologists use intravascular ultrasound (IVUS) for screening, risk assessment and stratification of coronary artery disease (CAD). We hypothesize that plaque components are vulnerable to rupture due to plaque progression. Currently, there are no standard grayscale IVUS tools for risk assessment of plaque rupture. This paper presents a novel strategy for risk stratification based on plaque morphology embedded with principal component analysis (PCA) for plaque feature dimensionality reduction and dominant feature selection technique. The risk assessment utilizes 56 grayscale coronary features in a machine learning framework while linking information from carotid and coronary plaque burdens due to their common genetic makeup. Method: This system consists of a machine learning paradigm which uses a support vector machine (SVM) combined with PCA for optimal and dominant coronary artery morphological feature extraction. Carotid artery proven intima-media thickness (cIMT) biomarker is adapted as a gold standard during the training phase of the machine learning system. For the performance evaluation, K-fold cross validation protocol is adapted with 20 trials per fold. For choosing the dominant features out of the 56 grayscale features, a polling strategy of PCA is adapted where the original value of the features is unaltered. Different protocols are designed for establishing the stability and reliability criteria of the coronary risk assessment system (cRAS). Results: Using the PCA-based machine learning paradigm and cross-validation protocol, a classification accuracy of 98.43% (AUC 0.98) with K = 10 folds using an SVM radial basis function (RBF) kernel was achieved. A reliability index of 97.32% and machine learning stability criteria of 5% were met for the cRAS. Conclusions: This is the first Computer aided design (CADx) system of its kind that is able to demonstrate the ability of coronary risk assessment and stratification while demonstrating a successful design of the machine learning system based on our assumptions.

Original languageEnglish (US)
Pages (from-to)137-158
Number of pages22
JournalComputer Methods and Programs in Biomedicine
Volume128
DOIs
StatePublished - May 1 2016

Fingerprint

Principal Component Analysis
Risk assessment
Principal component analysis
Learning systems
Coronary Artery Disease
Ultrasonics
Rupture
Support vector machines
Computer-Aided Design
Feature extraction
Endarterectomy
Carotid Arteries
Area Under Curve
Machine Learning
Coronary Vessels
Stability criteria
Biomarkers
Computer aided design
Screening
Planning

Keywords

  • Carotid IMT
  • Coronary artery
  • IVUS
  • Machine learning
  • PCA
  • Risk assessment

ASJC Scopus subject areas

  • Software
  • Computer Science Applications
  • Health Informatics

Cite this

PCA-based polling strategy in machine learning framework for coronary artery disease risk assessment in intravascular ultrasound : A link between carotid and coronary grayscale plaque morphology. / Araki, Tadashi; Ikeda, Nobutaka; Shukla, Devarshi; Jain, Pankaj K.; Londhe, Narendra D.; Shrivastava, Vimal K.; Banchhor, Sumit K.; Saba, Luca; Nicolaides, Andrew; Shafique, Shoaib; Laird, John R.; Suri, Jasjit S.

In: Computer Methods and Programs in Biomedicine, Vol. 128, 01.05.2016, p. 137-158.

Research output: Contribution to journalArticle

Araki, Tadashi ; Ikeda, Nobutaka ; Shukla, Devarshi ; Jain, Pankaj K. ; Londhe, Narendra D. ; Shrivastava, Vimal K. ; Banchhor, Sumit K. ; Saba, Luca ; Nicolaides, Andrew ; Shafique, Shoaib ; Laird, John R. ; Suri, Jasjit S. / PCA-based polling strategy in machine learning framework for coronary artery disease risk assessment in intravascular ultrasound : A link between carotid and coronary grayscale plaque morphology. In: Computer Methods and Programs in Biomedicine. 2016 ; Vol. 128. pp. 137-158.
@article{a7b9626aaff94eadaaaf3d0cde5616ee,
title = "PCA-based polling strategy in machine learning framework for coronary artery disease risk assessment in intravascular ultrasound: A link between carotid and coronary grayscale plaque morphology",
abstract = "Background and objective: Percutaneous coronary interventional procedures need advance planning prior to stenting or an endarterectomy. Cardiologists use intravascular ultrasound (IVUS) for screening, risk assessment and stratification of coronary artery disease (CAD). We hypothesize that plaque components are vulnerable to rupture due to plaque progression. Currently, there are no standard grayscale IVUS tools for risk assessment of plaque rupture. This paper presents a novel strategy for risk stratification based on plaque morphology embedded with principal component analysis (PCA) for plaque feature dimensionality reduction and dominant feature selection technique. The risk assessment utilizes 56 grayscale coronary features in a machine learning framework while linking information from carotid and coronary plaque burdens due to their common genetic makeup. Method: This system consists of a machine learning paradigm which uses a support vector machine (SVM) combined with PCA for optimal and dominant coronary artery morphological feature extraction. Carotid artery proven intima-media thickness (cIMT) biomarker is adapted as a gold standard during the training phase of the machine learning system. For the performance evaluation, K-fold cross validation protocol is adapted with 20 trials per fold. For choosing the dominant features out of the 56 grayscale features, a polling strategy of PCA is adapted where the original value of the features is unaltered. Different protocols are designed for establishing the stability and reliability criteria of the coronary risk assessment system (cRAS). Results: Using the PCA-based machine learning paradigm and cross-validation protocol, a classification accuracy of 98.43{\%} (AUC 0.98) with K = 10 folds using an SVM radial basis function (RBF) kernel was achieved. A reliability index of 97.32{\%} and machine learning stability criteria of 5{\%} were met for the cRAS. Conclusions: This is the first Computer aided design (CADx) system of its kind that is able to demonstrate the ability of coronary risk assessment and stratification while demonstrating a successful design of the machine learning system based on our assumptions.",
keywords = "Carotid IMT, Coronary artery, IVUS, Machine learning, PCA, Risk assessment",
author = "Tadashi Araki and Nobutaka Ikeda and Devarshi Shukla and Jain, {Pankaj K.} and Londhe, {Narendra D.} and Shrivastava, {Vimal K.} and Banchhor, {Sumit K.} and Luca Saba and Andrew Nicolaides and Shoaib Shafique and Laird, {John R.} and Suri, {Jasjit S.}",
year = "2016",
month = "5",
day = "1",
doi = "10.1016/j.cmpb.2016.02.004",
language = "English (US)",
volume = "128",
pages = "137--158",
journal = "Computer Methods and Programs in Biomedicine",
issn = "0169-2607",
publisher = "Elsevier Ireland Ltd",

}

TY - JOUR

T1 - PCA-based polling strategy in machine learning framework for coronary artery disease risk assessment in intravascular ultrasound

T2 - A link between carotid and coronary grayscale plaque morphology

AU - Araki, Tadashi

AU - Ikeda, Nobutaka

AU - Shukla, Devarshi

AU - Jain, Pankaj K.

AU - Londhe, Narendra D.

AU - Shrivastava, Vimal K.

AU - Banchhor, Sumit K.

AU - Saba, Luca

AU - Nicolaides, Andrew

AU - Shafique, Shoaib

AU - Laird, John R.

AU - Suri, Jasjit S.

PY - 2016/5/1

Y1 - 2016/5/1

N2 - Background and objective: Percutaneous coronary interventional procedures need advance planning prior to stenting or an endarterectomy. Cardiologists use intravascular ultrasound (IVUS) for screening, risk assessment and stratification of coronary artery disease (CAD). We hypothesize that plaque components are vulnerable to rupture due to plaque progression. Currently, there are no standard grayscale IVUS tools for risk assessment of plaque rupture. This paper presents a novel strategy for risk stratification based on plaque morphology embedded with principal component analysis (PCA) for plaque feature dimensionality reduction and dominant feature selection technique. The risk assessment utilizes 56 grayscale coronary features in a machine learning framework while linking information from carotid and coronary plaque burdens due to their common genetic makeup. Method: This system consists of a machine learning paradigm which uses a support vector machine (SVM) combined with PCA for optimal and dominant coronary artery morphological feature extraction. Carotid artery proven intima-media thickness (cIMT) biomarker is adapted as a gold standard during the training phase of the machine learning system. For the performance evaluation, K-fold cross validation protocol is adapted with 20 trials per fold. For choosing the dominant features out of the 56 grayscale features, a polling strategy of PCA is adapted where the original value of the features is unaltered. Different protocols are designed for establishing the stability and reliability criteria of the coronary risk assessment system (cRAS). Results: Using the PCA-based machine learning paradigm and cross-validation protocol, a classification accuracy of 98.43% (AUC 0.98) with K = 10 folds using an SVM radial basis function (RBF) kernel was achieved. A reliability index of 97.32% and machine learning stability criteria of 5% were met for the cRAS. Conclusions: This is the first Computer aided design (CADx) system of its kind that is able to demonstrate the ability of coronary risk assessment and stratification while demonstrating a successful design of the machine learning system based on our assumptions.

AB - Background and objective: Percutaneous coronary interventional procedures need advance planning prior to stenting or an endarterectomy. Cardiologists use intravascular ultrasound (IVUS) for screening, risk assessment and stratification of coronary artery disease (CAD). We hypothesize that plaque components are vulnerable to rupture due to plaque progression. Currently, there are no standard grayscale IVUS tools for risk assessment of plaque rupture. This paper presents a novel strategy for risk stratification based on plaque morphology embedded with principal component analysis (PCA) for plaque feature dimensionality reduction and dominant feature selection technique. The risk assessment utilizes 56 grayscale coronary features in a machine learning framework while linking information from carotid and coronary plaque burdens due to their common genetic makeup. Method: This system consists of a machine learning paradigm which uses a support vector machine (SVM) combined with PCA for optimal and dominant coronary artery morphological feature extraction. Carotid artery proven intima-media thickness (cIMT) biomarker is adapted as a gold standard during the training phase of the machine learning system. For the performance evaluation, K-fold cross validation protocol is adapted with 20 trials per fold. For choosing the dominant features out of the 56 grayscale features, a polling strategy of PCA is adapted where the original value of the features is unaltered. Different protocols are designed for establishing the stability and reliability criteria of the coronary risk assessment system (cRAS). Results: Using the PCA-based machine learning paradigm and cross-validation protocol, a classification accuracy of 98.43% (AUC 0.98) with K = 10 folds using an SVM radial basis function (RBF) kernel was achieved. A reliability index of 97.32% and machine learning stability criteria of 5% were met for the cRAS. Conclusions: This is the first Computer aided design (CADx) system of its kind that is able to demonstrate the ability of coronary risk assessment and stratification while demonstrating a successful design of the machine learning system based on our assumptions.

KW - Carotid IMT

KW - Coronary artery

KW - IVUS

KW - Machine learning

KW - PCA

KW - Risk assessment

UR - http://www.scopus.com/inward/record.url?scp=84960420615&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84960420615&partnerID=8YFLogxK

U2 - 10.1016/j.cmpb.2016.02.004

DO - 10.1016/j.cmpb.2016.02.004

M3 - Article

AN - SCOPUS:84960420615

VL - 128

SP - 137

EP - 158

JO - Computer Methods and Programs in Biomedicine

JF - Computer Methods and Programs in Biomedicine

SN - 0169-2607

ER -