Automated segmentation of biventricular contours in tissue phase mapping using deep learning

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 (V_x , V_y , V_z ) 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 (V_r and V_z ). 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 V_r and V_z 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 V_z and V_r 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.

On the influence of respiratory motion in radial tissue phase mapping cardiac MRI

PURPOSE

To investigate the impact of respiratory motion on radial tissue phase mapping (TPM) measurements, and to improve image quality and scan efficiency without compromising velocity fidelity by increasing the respiratory acceptance window with and without motion correction.

MATERIALS AND METHODS

A radial golden angle TPM sequence was measured in 10 healthy volunteers in three short axis slices at 3T. Ungated ( CFREE), self-gated with a single acceptance window ( CREF), motion-corrected averaging using all ( CMCall), or selected ( CMC) data reconstructions were compared by means of various image quality measures and resulting velocities.

RESULTS

Using all data ( CFREE) resulted in significantly higher perceived signal-to-noise ratio (SNR) (P < 0.001), but significantly reduced sharpness (P < 0.001) and contrast (P = 0.02), when compared to CREF. Coefficient of variation (CV) and perceived sharpness were not significantly different (P > 0.05). With motion-correction, perceived sharpness could be significantly improved ( CMC: P = 0.002; CMCall: P = 0.002) in comparison to CFREE. Velocity peaks of CFREE were significantly reduced compared to CREF (all peaks: P < 0.001; except the longitudinal "E" peak: P = 0.03). The peak velocities in CMC and CMCall were not significantly different from CREF (all peaks: P > 0.08; except longitudinal "E"/"A" peaks: P > 0.01).

CONCLUSION

Free-breathing reconstruction results in good perceived image sharpness and velocity information with slightly, but significantly, reduced peak velocities. For achieving velocities and image quality comparable to data from a single acceptance window, but higher gating efficiency, selected motion-corrected TPM (CMC) can be applied. J. Magn. Reson. Imaging 2016;44:1218-1228.