Automated segmentation of biventricular contours in tissue phase mapping using deep learning
- 2 September 2021
- journal article
- research article
- Published by Wiley in NMR in Biomedicine
- Vol. 34 (12) , e4606
Abstract
Tissue phase mapping (TPM) is an MRI technique for quantification of regional biventricular myocardial velocities. Despite its potential, clinical use is limited due to the requisite labor‐intensive manual segmentation of cardiac contours for all time frames. The purpose of this study was to develop a deep learning (DL) network for automated segmentation of TPM images, without significant loss in segmentation and myocardial velocity quantification accuracy compared with manual segmentation. We implemented a multi‐channel 3D (three dimensional; 2D + time) dense U‐Net that trained on magnitude and phase images and combined cross‐entropy, Dice, and Hausdorff distance loss terms to improve the segmentation accuracy and suppress unnatural boundaries. The dense U‐Net was trained and tested with 150 multi‐slice, multi‐phase TPM scans (114 scans for training, 36 for testing) from 99 heart transplant patients (44 females, 1‐4 scans/patient), where the magnitude and velocity‐encoded (Vx, Vy, Vz) images were used as input and the corresponding manual segmentation masks were used as reference. The accuracy of DL segmentation was evaluated using quantitative metrics (Dice scores, Hausdorff distance) and linear regression and Bland‐Altman analyses on the resulting peak radial and longitudinal velocities (Vr and Vz). The mean segmentation time was about 2 h per patient for manual and 1.9 ± 0.3 s for DL. Our network produced good accuracy (median Dice = 0.85 for left ventricle (LV), 0.64 for right ventricle (RV), Hausdorff distance = 3.17 pixels) compared with manual segmentation. Peak Vr and Vz measured from manual and DL segmentations were strongly correlated (R ≥ 0.88) and in good agreement with manual analysis (mean difference and limits of agreement for Vz and Vr were −0.05 ± 0.98 cm/s and −0.06 ± 1.18 cm/s for LV, and −0.21 ± 2.33 cm/s and 0.46 ± 4.00 cm/s for RV, respectively). The proposed multi‐channel 3D dense U‐Net was capable of reducing the segmentation time by 3,600‐fold, without significant loss in accuracy in tissue velocity measurements.Keywords
Funding Information
- American Heart Association (19IPLOI34760317)
- National Institutes of Health (R21EB030806, T32EB025766, R01HL117888, R01HL151079, R21AG055954, R21EB024315, R01HL138578, R01HL116895)
This publication has 48 references indexed in Scilit:
- Quantification of biventricular myocardial function using cardiac magnetic resonance feature tracking, endocardial border delineation and echocardiographic speckle tracking in patients with repaired tetralogy of fallot and healthy controlsJournal of Cardiovascular Magnetic Resonance, 2012
- The Utility of Cardiac Biomarkers, Tissue Velocity and Strain Imaging, and Cardiac Magnetic Resonance Imaging in Predicting Early Left Ventricular Dysfunction in Patients With Human Epidermal Growth Factor Receptor II–Positive Breast Cancer Treated With Adjuvant Trastuzumab TherapyJACC, 2011
- ACCF/ACR/AHA/NASCI/SCMR 2010 Expert Consensus Document on Cardiovascular Magnetic ResonanceJACC, 2010
- Comparison of myocardial velocities obtained with magnetic resonance phase velocity mapping and tissue doppler imaging in normal subjects and patients with left ventricular dyssynchronyJournal of Magnetic Resonance Imaging, 2006
- Assessment of Left Ventricular Dyssynchrony in Patients With Conduction Delay and Idiopathic Dilated CardiomyopathyJACC, 2006
- A fast and highly automated approach to myocardial motion analysis using phase contrast magnetic resonance imagingJournal of Magnetic Resonance Imaging, 2006
- Standardized Myocardial Segmentation and Nomenclature for Tomographic Imaging of the HeartCirculation, 2002
- A spatiotemporal model of cyclic kinematics and its application to analyzing nonrigid motion with MR velocity imagesIEEE Transactions on Medical Imaging, 1999
- Tracking myocardial deformation using phase contrast MR velocity fields: a stochastic approachIEEE Transactions on Medical Imaging, 1996
- Semiautomated method for noise reduction and background phase error correction in MR phase velocity dataJournal of Magnetic Resonance Imaging, 1993