ICM to Present Two Research Papers at MICCAI 2026
From 27 September to 1 October 2026, Strasbourg will host the MICCAI 2026 (International Conference on Medical Image Computing and Computer-Assisted Intervention), one of the leading international conferences on medical image processing and computer-assisted medicine, will take place in Strasbourg. The Institute for Cardiovascular Computer-Assisted Medicine (ICM) will be represented with two research papers in the fields of computer-assisted analysis of surgical image data and medical image analysis.
Processing multiple analysis tasks simultaneously
Julia Mellert will present the paper ‘Synergies and Conflicts in Multi-Task Surgical Vision: Stereo Matching, Segmentation, and Classification of Stereoendoscopic Video Frames in Mitral Valve Surgery”.
Die Arbeit beschäftigt sich mit der computergestützten Analyse von Videoaufnahmen während einer Mitralklappenoperation. Dabei werden drei unterschiedliche Analyseaufgaben miteinander kombiniert: die Berechnung von Tiefeninformationen aus stereoskopischen Aufnahmen, die Abgrenzung anatomischer Strukturen und die Klassifikation von Bildinhalten. Untersucht wird, inwieweit sich diese Aufgaben gegenseitig unterstützen und wo möglicherweise Zielkonflikte entstehen.
An der Arbeit sind neben Julia Mellert und Prof. Anja Hennemuth aus dem ICM unter anderem die DHZC-Wissenschaftler Dr. Julius Kaemmel, Prof. Jörg Kempfert und PD Dr. Simon Sündermann beteiligt.
Das Paper ist derzeit noch nicht öffentlich verfügbar.
3D and depth information for more precise image analysis
Also featured at MICCAI is the paper ‘Multimodal Mitral Valve Segmentation: Mesh-Based 3D and Depth-Transfer Learning’ by ICM researchers Julia Mellert, Lisa Bautz, Johanna Brosig and Prof. Anja Hennemuth, as well as Olaf Hellwich, Professor at the Technical University of Berlin. The work was carried out with the participation of the BIFOLD Agility Project AISUCO.
The focus is on the computer-aided delineation of the mitral valve in medical image data. To this end, three-dimensional, mesh-based models are combined with information on spatial depth. The aim is to make additional spatial information usable for machine learning and thereby improve the automatic analysis of anatomical structures.
The presentation will take place as part of the MVA specialist workshop on 1 October 2026.
With these two presentations, the ICM is showcasing its latest research on computer-aided analysis of surgical image data and three-dimensional medical image processing at a major international conference.
We wish our participating researchers every success in Strasbourg!
