TY - JOUR
T1 - Using intravascular ultrasound image-based fluid-structure interaction models and machine learning methods to predict human coronary plaque vulnerability change
AU - Wang, Liang
AU - Tang, Dalin
AU - Maehara, Akiko
AU - Wu, Zheyang
AU - Yang, Chun
AU - Muccigrosso, David
AU - Matsumura, Mitsuaki
AU - Zheng, Jie
AU - Bach, Richard
AU - Billiar, Kristen L.
AU - Stone, Gregg W.
AU - Mintz, Gary S.
N1 - Publisher Copyright:
© 2020, © 2020 Informa UK Limited, trading as Taylor & Francis Group.
PY - 2020
Y1 - 2020
N2 - Plaque vulnerability prediction is of great importance in cardiovascular research. In vivo follow-up intravascular ultrasound (IVUS) coronary plaque data were acquired from nine patients to construct fluid-structure interaction models to obtain plaque biomechanical conditions. Morphological plaque vulnerability index (MPVI) was defined to measure plaque vulnerability. The generalized linear mixed regression model (GLMM), support vector machine (SVM) and random forest (RF) were introduced to predict MPVI change (ΔMPVI = MPVIfollow-up‒MPVIbaseline) using ten risk factors at baseline. The combination of mean wall thickness, lumen area, plaque area, critical plaque wall stress, and MPVI was the best predictor using RF with the highest prediction accuracy 91.47%, compared to 90.78% from SVM, and 85.56% from GLMM. Machine learning method (RF) improved the prediction accuracy by 5.91% over that from GLMM. MPVI was the best single risk factor using both GLMM (82.09%) and RF (78.53%) while plaque area was the best using SVM (81.29%).
AB - Plaque vulnerability prediction is of great importance in cardiovascular research. In vivo follow-up intravascular ultrasound (IVUS) coronary plaque data were acquired from nine patients to construct fluid-structure interaction models to obtain plaque biomechanical conditions. Morphological plaque vulnerability index (MPVI) was defined to measure plaque vulnerability. The generalized linear mixed regression model (GLMM), support vector machine (SVM) and random forest (RF) were introduced to predict MPVI change (ΔMPVI = MPVIfollow-up‒MPVIbaseline) using ten risk factors at baseline. The combination of mean wall thickness, lumen area, plaque area, critical plaque wall stress, and MPVI was the best predictor using RF with the highest prediction accuracy 91.47%, compared to 90.78% from SVM, and 85.56% from GLMM. Machine learning method (RF) improved the prediction accuracy by 5.91% over that from GLMM. MPVI was the best single risk factor using both GLMM (82.09%) and RF (78.53%) while plaque area was the best using SVM (81.29%).
KW - Coronary artery disease
KW - fluid-structure interaction
KW - intravascular ultrasound
KW - machine learning
KW - plaque vulnerability prediction
UR - http://www.scopus.com/inward/record.url?scp=85088439731&partnerID=8YFLogxK
U2 - 10.1080/10255842.2020.1795838
DO - 10.1080/10255842.2020.1795838
M3 - Article
C2 - 32696674
AN - SCOPUS:85088439731
SN - 1025-5842
SP - 1267
EP - 1276
JO - Computer Methods in Biomechanics and Biomedical Engineering
JF - Computer Methods in Biomechanics and Biomedical Engineering
ER -