title:: Noninvasive hemodynamic assessment, treatment outcome prediction and follow-up of aortic coarctation from MR imaging
publisher:: American Association of Physicists in Medicine
people:: Kristóf Ralovich, Lucian Itu, Dime Vitanovski, Puneet Sharma, Razvan Ionasec, Viorel Mihalef, Waldemar Krawtschuk, Yefeng Zheng, Allen Everett, Giacomo Pongiglione, Benedetta Leonardi, Richard Ringel, Nassir Navab, Tobias Heimann, Dorin Comaniciu
organization:: Siemens Corporate Research, Technical University of Munich, Johns Hopkins University, Bambino Gesù Children’s Hospital, Transylvania University of Brașov, Friedrich-Alexander-Universität Erlangen-Nürnberg
domain:: Computational Fluid Dynamics, Medical Imaging, Clinical Applications, Cardiology
link:: https://doi.org/10.1118/1.4914856
citations:: 21 (OpenAlex) · 19 (Semantic Scholar) — as of 14 October 2026
Article record — Europe PMC.
Summary
The coarctation work taken all the way to a complete clinical workflow, and the first time three different settings are investigated across multiple subjects: preoperative severity assessment, post-stenting follow-up, and virtual stenting - predicting the treatment outcome before the intervention happens.
An end-to-end system of shape and computational models, personalised from MR imaging, computes the pressure drop across the coarctation non-invasively via CFD. Automatic thoracic aorta segmentation ran on 212 3D MR volumes at 8 seconds average. The ambition stated in the paper is explicit: given wider clinical validation, this non-invasive in-silico method could replace invasive pressure catheterisation.
Data Points
- First study to investigate three CoA settings on multiple subjects: preoperative severity, post-stenting follow-up, and virtual stenting for outcome prediction
- Automatic thoracic aorta segmentation on 212 3D MR volumes; mean symmetric point-to-mesh error 3.00 ± 1.58 mm, average 8 s
- Agreement with catheter measurement across 6 CoA patients: 2.38 ± 0.82 mmHg (preoperative), 1.10 ± 0.63 mmHg (post-stenting), 4.99 ± 3.00 mmHg (virtual stenting)
- Fast, mostly-automated workflow designed to be integrable into the clinical setting
- Virtual treatment method used to assess the predictive power of the approach, not just its descriptive accuracy
Reference
Medical Physics 42(5):2143-2156, 2015. doi:10.1118/1.4914856 · PubMed 25979009
19-21 citations (Semantic Scholar / OpenAlex, October 2026)
Part of the complete record in Scientific Publications - Peer-Reviewed Research Record.
