Effects of spatial and temporal resolution on cardiovascular magnetic resonance feature tracking measurements using a simple realistic numerical phantom

Q: What is the clinical evidence from this study?

Population: Healthy volunteer for cardiovascular magnetic resonance (CMR) numerical phantom development (n=1). Intervention: CMR feature tracking (FT) software packages (Medical Image Tracking Toolbox [MITT] and cvi42 v5.3.8 and v5.13.7) evaluated across varying in-plane spatial resolutions (1.4 x 1.4 mm2 to 3.0 x 3.0 mm2) and temporal resolutions (20 to 40 cardiac phases) vs. Ground-truth 2D Lagrangian radial and circumferential strain and displacement of the numerical phantom. Primary outcome: Difference from ground-truth peak radial and circumferential strain across spatial and temporal resolutions.

Q: What does this research mean for the field?

Increasing pixel size from 1.4 × 1.4 mm² to 3.0 × 3.0 mm² increases cardiovascular magnetic resonance feature tracking peak strain error relative to ground truth by up to 9.17 percentage points. Novelty: ClaimNovelty.CONFIRMATORY. Consensus alignment: ConsensusAlignment.SUPPORTS_CONSENSUS.

David M Adams; Redha Boubertakh; Marc E. Miquel

doi:10.1259/bjr.20220233

December 19, 2022British Journal of RadiologyOpen Access

Effects of spatial and temporal resolution on cardiovascular magnetic resonance feature tracking measurements using a simple realistic numerical phantom

View Full Paper

Ask AI

Bookmark

Share

Key result

Larger CMR pixel size increases peak radial strain error by ~9 percentage points in feature tracking.

n=1

Figures

Why the study?

The authors sought to develop a single-slice numerical phantom with known myocardial motion to test and compare cardiovascular magnetic resonance feature tracking software across varying temporal and in-plane spatial resolutions.

How do variations in in-plane spatial resolution, temporal resolution, and software version/vendor affect the accuracy of CMR feature tracking strain measurements compared to ground-truth strain in a numerical phantom?

Key Points

To evaluate the effects of in-plane spatial and temporal resolution on cardiovascular magnetic resonance feature tracking accuracy using a numerical phantom with known ground-truth motion.
Generated a single-slice numerical phantom with realistic texture and motion based on CMR acquisitions from one volunteer using the JEMRIS simulator.
Simulated images across multiple in-plane spatial resolutions (1.4 × 1.4 mm² to 3.0 × 3.0 mm²) and temporal resolutions (20 to 40 cardiac phases).
Quantified radial and circumferential strain and displacement against analytical ground truth using MITT and two versions of cvi42 (v5.3.8 and v5.13.7).
Larger pixel sizes increased divergence from ground truth, producing relative differences in peak strain of 9.17 percentage points (radial) and 8.42 percentage points (circumferential) between 1.4 × 1.4 mm² and 3.0 × 3.0 mm² for cvi42 v5.13.7.
Peak radial strain was undermeasured more by cvi42 v5.13.7 than v5.3.8, while MITT exhibited relative difference increases up to 203.44 percentage points between 2.0 × 2.0 mm² and 3.0 × 3.0 mm² for radial strain.
Temporal resolution showed no clear trend in relative peak strain error, yielding maximum changes between 20 and 25 phases of 1.43 percentage points for radial strain (cvi42 v5.3.8) and 1.67 percentage points for circumferential strain (cvi42 v5.13.7).

Structured PICO

How do variations in in-plane spatial resolution, temporal resolution, and software version/vendor affect the accuracy of CMR feature tracking strain measurements compared to ground-truth strain in a numerical phantom?

P

Population

1 healthy 33-year-old adult male volunteer whose CMR acquisitions (cine, tagging, T1/T2 maps, proton density) were used to construct a single-slice numerical phantom.

I

Intervention

Cardiovascular magnetic resonance feature tracking (CMR-FT) strain analysis using Medical Image Tracking Toolbox (MITT) and cvi42 (versions v5.3.8 and v5.13.7) across varying in-plane spatial resolutions (1.4 x 1.4, 2.0 x 2.0, 2.5 x 2.5, and 3.0 x 3.0 mm2 at 30 cardiac phases) and temporal resolutions (20, 25, 30, 35, and 40 cardiac phases corresponding to 61.65, 49.32, 41.10, 35.23, and 30.83 ms at 1.4 x 1.4 mm2).

C

Comparator

Known analytical ground-truth 2D Lagrangian radial and circumferential strain and myocardial displacement derived from tagging analysis (CIMTag2D) and non-rigid registration.

O

Outcome

Magnitude of relative and absolute differences in measured global peak radial and circumferential strain compared to ground-truth peak strain.surrogate

This numerical phantom study demonstrates that lower in-plane spatial resolution and software version updates substantially alter CMR feature tracking strain measurements relative to ground truth, highlighting the need for phantom-based validation after software updates.

Limitations

The numerical phantom lacks realism in certain aspects, specifically lacking through-plane motion, myocardial intensity changes through the cardiac cycle, and temporal variation of the blood-pool.
The numerical phantom was limited to only one mid-LV short-axis slice, restricting results to two-dimensional FT and omitting longitudinal strain.
The method of numerical phantom generation limited cardiac motion to that present in acquired images from a single volunteer rather than variable motion from electromechanical models.
Feature tracking software from only two vendors (MITT and cvi42) was evaluated.
The numerical phantom does not include through-plane motion, myocardial intensity changes through the cardiac cycle, or temporal variation of the blood-pool.
The phantom was limited to a single mid-LV short-axis slice, limiting evaluation to 2D feature tracking and excluding 3D feature tracking and longitudinal strain.
Cardiac motion was limited to that present in the acquired images of a single volunteer rather than variable cardiac motion generated by electromechanical models.
Feature tracking software of only two vendors was evaluated.

Ask about this paper

Answers are grounded in this paper and cite their sources.

Implication

Strain measurements from CMR feature tracking warrant caution at lower resolutions; leaves open the need for resolution-specific validation before routine use.

Abstract

Objectives To develop a single-slice numerical phantom with known myocardial motion, at several temporal and in-plane spatial resolutions, for testing and comparison of Cardiovascular Magnetic Resonance (CMR) feature tracking (FT) software. Methods The phantom was developed based on CMR acquisitions of one volunteer (acquired cine, tagging cine, T1 map, T2 map, proton density weighted image). The numerical MRI simulator JEMRIS was used, and the phantom was generated at several in-plane spatial resolutions (1.4 × 1.4 mm 2 to 3.0 × 3.0 mm 2 ) and temporal resolutions (20 to 40 cardiac phases). Two feature tracking software packages were tested: Medical Image Tracking Toolbox (MITT) and two versions of cvi42 (v5.3.8 and v5.13.7). The effect of resolution on strain results was investigated with reference to ground-truth radial and circumferential strain. Results Peak radial strain was consistently undermeasured more for cvi42 v5.13.7 than for v5.3.8. Increased pixel size produced a trend of increased difference from ground-truth peak strain, with the largest changes for cvi42 obtained using v5.13.7 between 1.4 × 1.4 mm 2 and 3.0 × 3.0 mm 2 , at 9.17 percentage points (radial) and 8.42 percentage points (circumferential). Conclusions The results corroborate the presence of intervendor differences in feature tracking results and show the magnitude of strain differences between software versions. Advances in knowledge This study shows how temporal and in-plane spatial resolution can affect feature tracking with reference to the ground-truth strain of a numerical phantom. Results reaffirm the need for numerical phantom development for the validation and testing of FT software.

Discussion

No takes yet. Share an insight, caveat, or question.

Cite This Study

Adams et al. (2022) studied Healthy volunteer for cardiovascular magnetic resonance (CMR) numerical phantom development (n=1). CMR feature tracking (FT) software packages (Medical Image Tracking Toolbox [MITT] and cvi42 v5.3.8 and v5.13.7) evaluated across varying in-plane spatial resolutions (1.4 x 1.4 mm2 to 3.0 x 3.0 mm2) and temporal resolutions (20 to 40 cardiac phases) vs. Ground-truth 2D Lagrangian radial and circumferential strain and displacement of the numerical phantom was evaluated on Difference from ground-truth peak radial and circumferential strain across spatial and temporal resolutions.

synapsesocial.com/papers/6a7f256fb6d9438a3a6a7a38 https://doi.org/10.1259/bjr.20220233

View Full Paper

Ask AI

Bookmark

Share

Effects of spatial and temporal resolution on cardiovascular magnetic resonance feature tracking measurements using a simple realistic numerical phantom

Figures

Key Points

Structured PICO

Limitations

Ask about this paper

Implication

Abstract

Discussion

Cite This Study

Effects of spatial and temporal resolution on cardiovascular magnetic resonance feature tracking measurements using a simple realistic numerical phantom

Figures

Key Points

Structured PICO

Limitations

Ask about this paper

Implication

Abstract

Discussion

Cite This Study

Keep reading

Keep reading