ws26 | Computational Imaging and AI in Medicine

AI-driven tools for Medical Imaging and Visualization (IN2107)

Fri, 19 Jun 2026 00:00:00 +0000

Modern healthcare is undergoing a digital transformation driven by the convergence of Artificial Intelligence (AI), medical imaging, and advanced visualization techniques. In particular, deep learning has demonstrated remarkable progress in medical image analysis, enabling improvements in image reconstruction and preprocessing, automated segmentation of anatomical and pathological structures, feature representation learning, and image visualization. These AI-driven imaging tools can have a profound impact on clinical workflows by enhancing diagnostic accuracy, facilitating image acquisition, enabling diagnostic standardization, and advancing medical training [1] [2].

Despite these advances, the clinical implementation of AI-based tools continues to face several challenges, including the standardization of workflows for image analysis, achieving model generalizability and fairness across patient populations and clinical sites [3], the fusion of multimodal information by combining different imaging modalities and clinical text, and the adoption of innovative visualization frameworks like augmented and mixed reality.

In this seminar, we will discuss recent advancements in medical image processing and their role in developing AI-assisted tools for clinical decision support and intervention. In particular, we will cover state-of-the-art techniques for biomedical image registration in space and time, fully automated and promptable segmentation models, and the use of virtual and augmented reality technologies to bring medical imaging data to life for surgical planning, medical training, and image-guided interventions. Furthermore, we will examine practical challenges associated with deploying these technologies in clinical settings and discuss their current and potential impact on modern healthcare practice.

References

[1] R. Najjar, “Redefining Radiology: A Review of Artificial Intelligence Integration in Medical Imaging,” Diagnostics, vol. 13, no. 17, p. 2760, Aug. 2023.

[2] A. Hosny, C. Parmar, J. Quackenbush, L. H. Schwartz, and H. J. W. L. Aerts, “Artificial intelligence in radiology,” Nature Reviews Cancer, vol. 18, no. 8, pp. 500–510, May 2018.

[3] Y. Yang, H. Zhang, J. W. Gichoya, D. Katabi, and M. Ghassemi, “The limits of fair medical imaging AI in real-world generalization,” Nature Medicine, vol. 30, no. 10, pp. 2838–2848, Jun. 2024.

Key topics to be covered include:

Traditional medical image visualization and navigation in the clinic
Image registration across modalities and time
Fully automated and promptable segmentation in medical imaging
3D rendering and virtual or augmented reality in medical imaging
State-of-the-art methods

Requirements:

Background in image processing and machine learning/deep learning
Interest in AI-tools for medical imaging
Interest in research

Please register via the TUM matching system: https://matching.in.tum.de

Check the intro slides here:

From General to Clinical Adapting Foundation Models for Medical Images (IN2107)

Thu, 05 Feb 2026 00:00:00 +0000

Course details

Foundation Models in Deep Learning for Medical Imaging Following a successful first iteration, this seminar extends the study of foundation models, with updated perspectives on adaptation strategies and applications in medical imaging. However, real-world deployment, especially in medical imaging, remains challenging due to unseen variations in scanners, patient demographics, and pathological conditions.

To address this issue, recent work has focused on foundation models: large-scale models pretrained on diverse data using self- or weakly supervised learning. CLIP-based models learn general-purpose representations that transfer well to downstream tasks. For the context of medical imaging, specific fine-tuned models such as MedSAM and BiomedCLIP demonstrate strong potential.

However, adapting foundation models to new domains or tasks often remains challenging. Although domain generalization and adaptation approaches exist, they may overfit or require target data. A recent paradigm, test-time adaptation, mitigates this by adapting alongside inference to unseen data. The techniques include adapting dynamically through model updates or prompt modifications that support more flexible deployment.

This seminar will explore the foundations and capabilities of foundation models, with a focus on pretraining strategies, representational generalization, and efficient adaptation methods. We will examine state-of-the-art foundation models and discuss how they can be adapted to downstream tasks, especially in medical imaging, through techniques such as self-supervised learning, cross-domain transfer, and parameter-efficient tuning.

Key topics to be covered include:

Introduction to Self Supervised Learning and Foundation Models
Adaptation techniques to clinical applications
Examples of Foundation models in medical imaging
State-of-the-art methods

Requirements:

Background in image processing and machine learning/deep learning
Interest in medical image analysis and foundation models
Interest in research

Please register via the TUM matching system: https://matching.in.tum.de

Check the intro slides here:

Multimodal Learning for Medicine & Healthcare The Challenges and Opportunities (IN2107)

Thu, 05 Feb 2026 00:00:00 +0000

Course details

Doctors typically make clinical decisions using several modalities, such as images, language, or tabular data. Deep learning models offer a powerful framework for integrating these heterogeneous modalities to support automated and data-driven decision making. However, effective multimodal learning remains non-trivial, facing challenges such as under-optimization across modalities [1] or the presence of missing data [2]. These challenges are particularly pronounced in real-world clinical settings, where data availability, quality, and alignment across modalities vary substantially. In this seminar, we will discuss recent advances in multimodal learning, covering key paradigms such as fusion and alignment mechanisms, self-supervised and contrastive pretraining across modalities, and the emergence of multimodal foundation models for medical AI. We will also examine strategies that address real-world challenges, including handling missing or noisy modalities, improving cross-modal generalization, and enhancing data efficiency and robustness for clinical applications.

References

[1] Shicai Wei, Chunbo Luo, and Yang Luo. Boosting multimodal learning via disentangled gradient learning. arXiv preprint arXiv:2507.10213, 2025.

[2] Sijie Li, Chen Chen, and Jungong Han. Simmlm: A simple framework for multi-modal learning with missing modality. arXiv preprint arXiv:2507.19264, 2025.

[3] Wu, Zhenbang, et al. “Multimodal patient representation learning with missing modalities and labels.” The Twelfth International Conference on Learning Representations. 2024.

[4] Yun, Sukwon, et al. “Flex-moe: Modeling arbitrary modality combination via the flexible mixture-of-experts.” Advances in Neural Information Processing Systems 37 (2024): 98782-98805.

[5] Zhang, Kai, et al. “A generalist vision–language foundation model for diverse biomedical tasks.” Nature Medicine 30.11 (2024): 3129-3141.

[6] Radford, Alec, et al. “Learning transferable visual models from natural language supervision.” International conference on machine learning. PmLR, 2021.

[7] Ma, Jun, et al. “Segment anything in medical images.” Nature Communications 15.1 (2024): 654.

[8] Li, Songtao, and Hao Tang. “Multimodal alignment and fusion: A survey.” arXiv preprint arXiv:2411.17040 (2024).

Key topics to be covered include:

Introduction to multimodal learning in medicine
Challenges of multimodal learning in clinical applications, including missing and noisy data
Multimodal pretraining for medicine
State-of-the-art methods

Requirements:

Background in image processing and machine learning/deep learning
Interest in medical multimodal learning
Interest in research

Please register via the TUM matching system: https://matching.in.tum.de

Check the intro slides here:

Artificial Intelligence in Medicine (IN2403)

Fri, 01 Oct 2021 00:00:00 +0000

At the end of the module students should be able to recall the important topics in the area of artificial intelligence in medicine, understand the relations between the topics, apply their knowledge to own deep learning projects, analyse and evaluate social and ethical implications and develop own strategies to apply the learned concepts to their own work.

Introduction: Clinical motivation, clinical data, clinical workflows
ML for medical imaging• Data curation for medical applications
Domain shift in medical applications: Adversarial learning and Transfer learning
Self-supervised learning and unsupervised learning
Learning from sparse and noisy data
ML for unstructured and multi-modal clinical data
NLP for clinical data• Bayesian approaches to deep learning and uncertainty
Interpretability and explainability
Federated learning, privacy-preserving ML and ethics
ML for time-to-event modeling, survival models
ML for differential diagnosis and stratification• Clinical applications in pathology/radiology/omics