Artificial intelligence and computer-aided support for diagnosis and therapy in digital medicine—these are the topics the Fraunhofer Institute for Digital Medicine MEVIS is presenting at the SPIE Medical Imaging conference, which is held in San Diego/USA from February 16 to 20. This year, Fraunhofer MEVIS is again well represented at the renowned conference with numerous scientific and educational contributions, including a live demonstration workshop, two training courses, three oral and six poster presentations.

Institute Director Horst Hahn will join forces with Karen Drukker from University of Chicago Medicine and Lubomir M. Hadjiiski from Michigan Medicine to chair the Live Demonstrations Workshop. The workshop provides a forum for innovators to showcase their creations through live system demonstrations and software presentations. The primary objectives are to inspire research, offer valuable feedback to participants, foster new collaborations, and facilitate learning in the realm of medical imaging.
February 18, 5:30 pm – 7:00 pm PST, Golden State Ballroom

Training Courses

• Fraunhofer MEVIS scientists Hans Meine and Felix Thielke will conduct a full-day course on Transformers: A powerful tool for Medical Image Analysis and Generation.
  February 16, 8:30 am – 5:30 pm PST, Course SC1324
• Their colleagues Markus Wenzel and Eike Petersen will offer a half-day course on AI-based Medical Imaging: Fairness in Models and Causality in Images.
  February 17, 1:30 pm – 5:30 pm PST, Course SC1351

Both hands-on courses are aiming at students, researchers, and engineers from academia and industry who want to expand and deepen their knowledge of state-of-the-art architectures and AI models.

In addition, Fraunhofer MEVIS will contribute with the following scientific presentations at SPIE Medical Imaging 2025:

Oral Presentations

• Feature extractor sensitivity in synthetic medical image evaluation, Daniel Mensing et al.
  February 18, 11:50 am – 12:10 pm PST, Town & Country A
• Deep Learning-Based Outcome Prediction from Endoscopic Videos for Robot-Assisted Radical Prostatectomy, Chiara Tappermann et al.
  February 18, 3:40 pm – 4:00 pm PST, Town & Country A
• Left ventricle mesh generation for deformation analysis based on 3D echocardiographic images, Lisa Bautz et al.
  February 20, 2:30 pm – 2:50 pm PST, Palm 6

Posters

• Landmark based 2D/3D registration of pre-operative planning data to intra-operative photographs for mapping of functional areas in awake brain surgery, Florian Weiler et al.
  February 17, 5:30 pm – 7:00 pm PST, Golden State Ballroom
• Predicting Anatomical Structures from 2D Depth Images of Patients, Kai Geißler et al.
  February 19, 5:30 pm – 7:00 pm PST, Golden State Ballroom
• Multi-site segmentation of breast and fibroglandular tissue in MRI with a focus on clinical practicality, Kai Geißler et al.
  February 19, 5:30 pm – 7:00 pm PST, Golden State Ballroom
• MR-to-CT synthesis for cross-modality model adaptation, Daniel Mesing et al.
  February 19, 5:30 pm – 7:00 pm PST, Golden State Ballroom
• Pitfalls with anomaly detection for breast cancer risk prediction, Tom Koller et al.
  February 19, 5:30 pm – 7:00 pm PST, Golden State Ballroom
• Breast cancer risk prediction using background parenchymal enhancement, radiomics, and symmetry features on MRI, Kai Geißler et al.
  February 19, 5:30 pm – 7:00 pm PST, Golden State Ballroom