Assessment of infarct-specific cardiac motion dysfunction using modeling and multimodal magnetic resonance merging

Chen Onn Leong; Yih Miin Liew; Mehmet Bilgen; Yang Faridah Abdul Aziz; Kok Han Chee; Yin Kia Chiam; Einly Lim

doi:10.1002/jmri.25390

Assessment of infarct-specific cardiac motion dysfunction using modeling and multimodal magnetic resonance merging

J Magn Reson Imaging. 2017 Feb;45(2):525-534. doi: 10.1002/jmri.25390. Epub 2016 Jul 15.

Authors

Chen Onn Leong¹, Yih Miin Liew¹, Mehmet Bilgen², Yang Faridah Abdul Aziz³, Kok Han Chee⁴, Yin Kia Chiam⁵, Einly Lim¹

Affiliations

¹ Department of Biomedical Engineering, Faculty of Engineering, University of Malaya, Kuala Lumpur, Malaysia.
² Biophysics Department, Faculty of Medicine, Adnan Menderes University, Aydin, Turkey.
³ Department of Biomedical Imaging, University Malaya Research Imaging Centre, Faculty of Medicine, University of Malaya, Kuala Lumpur, Malaysia.
⁴ Department of Medicine, Faculty of Medicine, University of Malaya, Kuala Lumpur, Malaysia.
⁵ Department of Software Engineering, Faculty of Computer Science & Information Technology, University of Malaya, Kuala Lumpur, Malaysia.

PMID: 27418150
DOI: 10.1002/jmri.25390

Abstract

Purpose: To propose a cardiac motion tracking model that evaluates wall motion abnormality in postmyocardial infarction patients. Correlation between the motion parameter of the model and left ventricle (LV) function was also determined.

Materials and methods: Twelve male patients with post-ST elevation myocardial infarction (post-STEMI) and 10 healthy controls of the same gender were recruited to undergo cardiac magnetic resonance imaging (MRI) using a 1.5T scanner. Using an infarct-specific LV division approach, the late gadolinium enhancement (LGE) MRI images were used to divide the LV on the tagged MRI images into infarct, adjacent, and remote sectors. Motion tracking was performed using the infarct-specific two-parameter empirical deformable model (TPEDM). The match quality was defined as the position error computed using root-mean-square (RMS) distance between the estimated and expert-verified tag intersections. The position errors were compared with the ones from our previously published fixed-sector TPEDM. Cine MRI images were used to calculate regional ejection fraction (REF). Correlation between the end-systolic contraction parameter (α_ES ) with REF was determined.

Results: The position errors in the proposed model were significantly lower than the fixed-sector model (P < 0.01). The median position errors were 0.82 mm versus 1.23 mm. The α_ES correlates significantly with REF (r = 0.91, P < 0.01).

Conclusion: The infarct-specific TPEDM combines the morphological and functional information from LGE and tagged MRI images. It was shown to outperform the fixed-sector model in assessing regional LV dysfunction. The significant correlation between α_ES and REF added prognostic value because it indicated an impairment of cardiac function with the increase of infarct transmurality.

Level of evidence: 3 J. Magn. Reson. Imaging 2017;45:525-534.

Keywords: LV motion analysis; cardiac modeling; late gadolinium enhancement; myocardial infarction; tagged MRI.

Publication types

Research Support, Non-U.S. Gov't

MeSH terms

Computer Simulation
Humans
Image Interpretation, Computer-Assisted / methods
Magnetic Resonance Imaging, Cine / methods*
Male
Middle Aged
Models, Cardiovascular*
Motion
Movement
Multimodal Imaging / methods
Myocardial Infarction / complications
Myocardial Infarction / diagnostic imaging*
Myocardial Infarction / physiopathology*
Reproducibility of Results
Sensitivity and Specificity
Subtraction Technique*
Ventricular Dysfunction, Left / diagnostic imaging*
Ventricular Dysfunction, Left / etiology
Ventricular Dysfunction, Left / physiopathology*