The project
With university affiliation through the DBMR, we develop multimodal AI for whole-body PET/CT, integrating imaging with clinical context to distinguish residual malignancy from inflammation or physiological uptake. The focus is lymphoma response assessment, with methods designed to transfer across tasks, diseases and institutions. You will work with over 14,000 available PET/CT examinations from Bern (>30 TB), linked reports, referrals and longitudinal studies, multicentre trial cohorts, and digitised histology in selected patients.
The project
With university affiliation through the DBMR, we develop multimodal AI for whole-body PET/CT, integrating imaging with clinical context to distinguish residual malignancy from inflammation or physiological uptake. The focus is lymphoma response assessment, with methods designed to transfer across tasks, diseases and institutions. You will work with over 14,000 available PET/CT examinations from Bern (>30 TB), linked reports, referrals and longitudinal studies, multicentre trial cohorts, and digitised histology in selected patients.
Your Responsibilities
- Shape research questions and develop multimodal methods combining PET/CT with clinical text, followed by longitudinal imaging and digitised histology.
- Develop context-aware image analysis and segmentation, multimodal representation learning and outcome prediction; explore physician-editable report generation.
- Lead projects from model design through multicentre clinical validation.
- Aim to publish in leading international journals and present at machine-learning and medical-imaging conferences.
- Contribute to follow-on grant applications and interdisciplinary collaboration.
Your profile
- PhD in computer science, machine learning, biomedical engineering or a related field.
- Strong deep-learning research on images; 3D/volumetric experience highly desirable.
- Excellent Python/PyTorch skills; reproducible pipelines at scale.
- Substantial methodological ownership and strong first-author publications.
- Ability to frame research questions, design baselines/ablations, identify leakage and shortcut learning, and communicate across disciplines.
- Particularly valuable: multimodal/vision-language learning, self-supervised pretraining or medical foundation models, longitudinal imaging, clinical NLP/report generation, computational pathology/whole-slide imaging, and distributed training.
- Prior PET/CT experience is not required; we provide clinical, biological and imaging expertise.
Your Benefits
- Meaningful work and fair compensation
- Strong research infrastructure and international network
- Interdisciplinary collaboration
- Access to cutting-edge technology and innovation
#J-18808-Ljbffr
Postdoctoral Researcher - Multimodal AI for Whole-Body PET/CT in Bern Arbeitgeber: Kanton Bern
Die Universität Bern ist ein hervorragender Arbeitgeber, der eine inspirierende Arbeitsumgebung bietet, in der Teamarbeit und persönliche Entwicklung gefördert werden. Als Teamleiter im Bereich Administration & Debitoren-Management profitieren Sie von flexiblen Arbeitszeiten, umfangreichen Weiterbildungsmöglichkeiten und einer wertschätzenden Unternehmenskultur, die Vielfalt und Innovation schätzt. Die zentrale Lage der Universität ermöglicht zudem eine gute Erreichbarkeit und ein lebendiges Umfeld für alle Mitarbeitenden.
