MR

Team: Agnes Mayr, Felix Troger, Christian Kremser, Peter Schullian, Matthias Schwab, Enrique Almar Munoz

The "Cardiovascular MRI” research group for many years works in close cooperation with the University Clinic for Internal Medicine III/Cardiology and Angiology (Prof. Dr. B. Metzler, Ass.-Prof. PD Dr. SJ. Reinstadler, PhD) with the main focus targeting the optimized stratification of patients prognosis after ST-elevation myocardial infarction based by cardiac MRI parameters. In addition, the development, randomized validation and AI-supported automation of an MRI-based TAVI planning alternative (transcatheter aortic valve intervention) represents a current research approach. Other cooperation projects include the shock wave therapy after myocardial infarction with the University Hospital for Cardiac Surgery Innsbruck (CAST-HF study , PD Dr. J. Holfeld).
Together with these local and international project partners, the cardiovascular MRI research group was able to publish more than 100 publications over the last years.

selected publications:
Troger F, Klug G, Poskaite P, Tiller C, Lechner I, Reindl M, Holzknecht M, Fink P, Brunnauer E, Gizewski E, Metzler B, Reinstadler S, Mayr A. Mitral Annular Disjunction in Out-of-Hospital Cardiac Arrest Patients - a Retrospective Cardiac MRI Study. Clin Res Cardiol. 113(5):770-780 (2024)

Reindl M, Lechner I, Holzknecht M, Tiller C, Fink P, Oberhollenzer F, von der Emde S, Pamminger M, Troger F, Kremser C, Laßnig E, Danninger K, Binder RK, Ulmer H, Brenner C, Klug G, Bauer A, Metzler B, Mayr A*, Reinstadler SJ*. Cardiac Magnetic Resonance Imaging Versus Computed Tomography to Guide Transcatheter Aortic Valve Replacement (TAVR-CMR): A Randomized, Open-Label, Non-Inferiority Trial. Circulation. 148:1220-1230 (2023).

Mayr A, Klug G, Reindl M, Lechner I, Tiller C, Holzknecht M, Pamminger M, Troger F, Schocke M, Bauer A, Reinstadler SJ, Metzler M. Evolution of myocardial tissue injury and persistence of iron and edema within the infarct core: a CMR study over a decade after ST-elevation myocardial infarction. JACC Cardiovasc Imaging. 15(6):1030-1042 (2022).

Pamminger M, Reindl M, Kranewitter C, Troger F, Tiller C, Holzknecht M, Lechner I, Poskaite P, Klug G, Kremser C, Reinstadler SJ, Metzler B, Mayr A. Prognostic value of pulmonary transit time by cardiac magnetic resonance imaging in ST-elevation myocardial infarction. Eur Radiol. 33(2):1219-1228 (2023).

Holzknecht M, Reindl M, Tiller C, Reinstadler SJ, Lechner I, Pamminger M, Schwaiger JP, Klug G, Bauer A, Metzler B, Mayr A. Global longitudinal strain improves risk assessment after ST-segment elevation myocardial infarction: a comparative prognostic evaluation of left ventricular functional parameters. Clin Res Cardiol 110, 1599-1611 (2021).

Mayr A, Klug G, Reinstadler SJ, Feistritzer HJ, Reindl M, Kremser C, Kranewitter C, Bonaros N, Friedrich G, Feuchtner G, Metzler B (2018): Is MRI equivalent to CT in the guidance of TAVR? A pilot study. Eur Radiol 28, 4625-4634 (2018).

Development of artificial intelligence (AI) methods for the automatic analysis of Cardiovascular Magnetic Resonance (CMR) images following myocardial infarction: in particular, algorithms for measuring infarct size and detecting infarct complications (such as microvascular obstruction and intramyocardial hemorrhage), as well as comparisons of AI-based measurements with traditional manual measurements and evaluation of their correlation with clinical biomarkers and patient outcomes to better understand how reliable AI-based quantification is compared to human experts and whether it can be trusted in future clinical practice.

Selected publications:
Schwab M, Pamminger M, Kremser C, Obmann D, Haltmeier M, Mayr A. Error correcting 2D-3D cascaded network for myocardial infarct scar segmentation on late gadolinium enhancement cardiac magnetic resonance images. Medical Image Analysis. 2025;103:103594. doi:https://doi.org/10.1016/j.media.2025.103594

Schwab M, Pamminger M, Kremser C, Haltmeier M, Mayr A. Deep learning pipeline for fully automated myocardial infarct segmentation from clinical cardiac MR scans. Radiology Advances. 2025;2(4):umaf023. doi:https://doi.org/10.1093/radadv/umaf023.

Schwab M, Mayr A, Haltmeier M. Deep gaussian mixture model for unsupervised image segmentation. In: Nicosia G, Ojha V, Giesselbach S, Pardalos MP, Umeton R, editors. Machine Learning, Optimization, and Data Science. Cham: Springer Nature Switzerland; 2025. p. 339-52. doi:https://doi.org/10.1007/978-3-031-82484-5_25.

Schwab M, Haltmeier M, Mayr A. Disagreement-Driven Uncertainty Quantification in Late Gadolinium Enhancement Cardiac MRI. In: Sudre CH, Hoque MI, Mehta R, Ouyang C, Qin C, Rakic M, et al., editors. Uncertainty for Safe Utilization of Machine Learning in Medical Imaging. Cham: Springer Nature Switzerland; 2026. p. 24-33. doi:https://doi.org/10.1007/978-3-032-06593-3_3.

Application of AI methods for planning transcatheter aortic valve implantation (TAVI) using CMR imaging: automatic identification of important TAVI structures and estimation of the severity of heart valve calcification on CMR images, prediction of possible surgical complications and survival time after TAVI based on CMR images.

Selected publications:
Colin-Tenorio, C. G., Mayr, A., Kremser, C., Haltmeier, M., & Almar-Munoz, E. (2025). Cardiac Magnetic Resonance-to-Computed Tomography Angiography image conversion using diffusion models for Transcatheter Aortic Valve Implantation planning. Intelligence Based Medicine, 100335

Almar-Munoz, E., Pamminger, M., Kremser, C., Haltmeier, M., \& Mayr, A. (2024, October). Beyond the standards: Fully-automated aortic annulus segmentation on contrast-free magnetic resonance imaging using a computational aorta unwrapping method. In International Workshop on Statistical Atlases and Computational Models of the Heart (pp. 110-121). Cham: Springer Nature Switzerland.

Almar-Munoz, E., Tiefenthaler, M., Leon Contreras, N. S., Aguilar-Minguez, A., Kremser, C., Haltmeier, M., & Mayr, A. (2024, October). Multi-loss 3D Segmentation for Enhanced Bi-atrial Segmentation. In International Workshop on Statistical Atlases and Computational Models of the Heart (pp. 236-244). Cham: Springer Nature Switzerland.

Islam, M., Tabassum, M., Mayr, A., Kremser, C., Haltmeier, M., & Almar-Munoz, E. (2025). Uncertainty-Guided Active Learning for Access Route Segmentation and Planning in Transcatheter Aortic Valve Implantation. Journal of Imaging, 11(9), 318.