Seminar · Medical Faculty
Artificial Intelligence in Radiology: Applications and Research (AIR)
A seminar for medical students on how deep learning is used in radiology: classification, detection, segmentation, reconstruction, tracking and disease progression, together with bias, uncertainty, foundation models and the ethics and regulation of AI in healthcare.
About the course
The Albarqouni Lab at the Clinic for Diagnostic and Interventional Radiology, University Hospital Bonn, offers this seminar every semester. Over two semester hours per week (2 SWS), medical students explore how deep learning is transforming medical imaging, from the foundations of neural networks to classification, detection, segmentation, reconstruction, tracking and disease progression, and on to current topics such as causality and bias, uncertainty, foundation and vision-language models, and the ethical and regulatory side of clinical AI.
In BASIS the seminar is currently listed as AI and Medical Imaging (no. 401040415). It has also run as Seminar on AI and Medical Imaging (no. 401040137, WS 2022/23 – SS 2025) and under its German title (no. 401040432), and in WS 2025/26 as a joint special edition with the German University in Cairo. Each semester page gives the title and number it was listed under.
Course structure
| Session | Lecture and topics |
|---|---|
| Week 1 · kick-off | L1 Why AI in medicine clinical pressure points; promise and evidence; what goes wrong; humans and AI together |
| Week 2 | L5 Deep learning for classification and detection (primer) perceptron to CNN; architectures and transfer; detection, IoU, mAP; class imbalance |
| Week 4 | L4 Evaluation with clinical metrics (primer) confusion matrix; ROC and PR; prevalence and operating points; confidence intervals |
| Week 6 | L7 Segmentation (primer) classic methods; U-Net and nnU-Net; Dice, HD95, losses; few labels and SAM; registration and tracking (rigid to deformable, mutual information, lesion and cell tracking) |
| Week 8 | L11 Uncertainty quantification (primer) aleatoric and epistemic; MC dropout and ensembles; calibration; refer or decide; conformal prediction; uncertainty for graphs and LLMs |
| Week 10 | L12 Bias, fairness and causality (primer) causal diagrams; dataset shift; fairness metrics; mitigation; bias in graphs and foundation models |
| Week 12 | L9 Foundation models and multimodal data (primer) self-supervision; multimodal fusion and contrastive learning; vision–language models (CLIP); clinical LLMs and agents |
| Week 13 | L13 Trustworthy AI in practice (primer) ethics and the EU AI Act; evidence and reporting; monitoring and drift; federated learning and privacy; carbon footprint |
Toolbox
- T1 Vectors, matrices and gradients: scalars to tensors; dot products and linear maps; distances, Euclidean to geodesic; convexity; gradient, Jacobian, Hessian
- T2 Probability and statistics: z-scores and outliers; distributions, Bernoulli to Rician; CLT and confidence intervals; Bayes and Bayesian updating; MLE, MAP, entropy, KL
- T3 Python toolbox: notebooks and Colab; NumPy on the course CT; pandas; scikit-learn and PyTorch cheat sheets · Lab: arrays, tables, models
Learning objectives
- Explain deep-learning concepts and how they are applied in medical imaging.
- Analyse advanced image-analysis tasks, including classification, detection, segmentation and disease-progression modelling.
- Evaluate ethical and regulatory considerations in deploying AI in healthcare.
- Discuss the future of AI-driven solutions in personalised medicine and patient care.
Target audience
Teaching format
Prerequisites
- Familiarity with fundamental medical imaging techniques and an interest in AI applications.
- Desirable: an understanding of medical diagnostics and clinical data interpretation.
Materials
- A reading list of review papers for each session, linked in the semester schedule.
- Slides of the participants’ presentations, shared on eCampus.
Evaluation
- Graded seminar: the grade is based on the paper presentation and active participation in the discussion.
Semesters
| Semester | Course | Format | Audience | Lecturers & tutors |
|---|---|---|---|---|
| SS 2027 | Artificial Intelligence in Radiology: Applications and Research (AIR) | Seminar | Medical Faculty | |
| WS 2026/27 | Artificial Intelligence in Radiology: Applications and Research (AIR) listed as: AI and Medical Imaging, no. 401040415 | Seminar | Medical Faculty | |
| SS 2026 | Artificial Intelligence in Radiology: Applications and Research (AIR) | Seminar | Medical Faculty | |
| WS 2025/26 | Artificial Intelligence in Radiology: Applications and Research (AIR) listed as: Artifizielle Intelligenz in der Radiologie: Anwendungen und Forschung (special edition with the German University in Cairo) | Seminar | Medical Faculty | |
| SS 2025 | Artificial Intelligence in Radiology: Applications and Research (AIR) listed as: Seminar on AI and Medical Imaging, no. 401040137 | Seminar | Medical Faculty | |
| WS 2024/25 | Artificial Intelligence in Radiology: Applications and Research (AIR) | Seminar | Medical Faculty | |
| WS 2023/24 | Artificial Intelligence in Radiology: Applications and Research (AIR) listed as: Seminar on AI and Medical Imaging, no. 401040137 | Seminar | Medical Faculty | |
| SS 2023 | Artificial Intelligence in Radiology: Applications and Research (AIR) listed as: Seminar on AI and Medical Imaging | Seminar | Medical Faculty | |
| WS 2022/23 | Artificial Intelligence in Radiology: Applications and Research (AIR) listed as: Seminar on AI and Medical Imaging | Seminar | Medical Faculty |