Introduction
Late gadolinium enhancement (LGE) MRI is known as the gold standard for identification of post-infarction tissue alterations in the myocardium. However, restrictions may limit the use of contrast agents in some patients. Ultra-high field (UHF) cardiac MRI (B0 ≥ 7T) is particularly interesting due to the increased resolution, sensitivity and ability to explore new contrast mechanisms. Due to high technical requirements, especially with regard to SAR safety, LGE-measurements for humans at UHF are restricted.  The potential of T₂* contrast at 7T to visualize cardiac tissue alterations in acute and subacute stages after myocardial infarction (MI) has been presented previously¹.
This study aimed to statistically analyze and compare the areas of altered contrast in the T₂*-weighted (T₂*w) and LGE measurements performed in a pig model of myocardial infarction.

Methods
After approval by the Animal Welfare Committee, myocardial infarction of the LAD with following reperfusion after 90 minutes was induced in seven German Landrace Pigs. At the Magnetom™ "Terra" 7T MR scanner (Siemens, Erlangen) four MRI measurements were performed using in-house-developed weight-matched 8Tx/16Rx cardiac arrays. The analysis of areas with visible contrast changes included two MR measurements (3-4 and 10-14 days after MI) from three pigs with a total of 42 short-axis slices of corresponding T₂*w images (TE 4.83 ms, voxel size 2.2x2.5x6 mm^₃) and LGE images (voxel size 0.7x0.7x6 mm³). Using FiJi software injured areas were labeled separately based on optical contrast after preliminary contouring of epicardial and endocardial borders. An example of images in midcavity position is shown in Fig. 1a. Statistical analysis of the labeled area size, including a Pearson correlation and paired t-test, was performed using Origin Pro 9.

Results
On both T₂*w and LGE images, the altered contrast areas are clearly visible in the septum and anterior wall of the left ventricle. Prominent hypointense inhomogeneities are seen in areas of presumed reperfusion tissue damage on T₂*w images, whereas hyperintense contrast enhancement is seen on LGE images (Fig. 1a).
The Pearson correlation of the altered contrast area size in LGE and T₂* shows a strong correlation for all slices. When differentiating between slice positions, a significantly better correlation is evident in basal/midcavity slices (Fig. 1b).
Comparison of the total size of the altered contrast area in LGE and T₂*w images shows a significant difference in the size of the measured area in a paired t-test (Fig. 2). 

Discussion
We demonstrate a substantial co-localization of position and a good correlation between the sizes of the altered contrast areas in T₂*w and LGE images in the early phase of tissue alteration after myocardial infarction especially in basal and midcavity slice positions. The absolute amount of altered image intensity is statistically significantly smaller in the T₂*w images than in the LGE images, which could reflect presumed physiological differences in the background of the two contrasts.
Both measurement and evaluation methods for  7T LGE and T₂* data at apical regions await further technical improvements.

Conclusion
T₂*w 7T MR images may become a method to detect and quantify acute and subacute postinfarction myocardial injury without the need for contrast agent application.

Acknowledgements
BMBF grants: 01EO1004, 01E1O1504

¹Aures et al. (2022). DGK Jahrestagung 2022. Abstract 362