PhD Position (100%, E13; f/m/d): Multimodal Embeddings and Fusion Architectures for Digital Patient Twins

TUM - PhD Position (100%, E13; f/m/d): Multimodal Embeddings and Fusion Architectures for Digital Patient Twins

Full-time PhD Student (f/m/d) for the EU research project TWIN-X

PhD Position (100%, E13; f/m/d): Multimodal Embeddings and Fusion Architectures for Digital Patient Twins

04.06.2026,Academic staff

We are seeking a full-time PhD Student (f/m/d) for the EU project TWIN-X at TUM University Hospital. The position focuses on multimodal embeddings and fusion architectures for digital patient twins in oncology and cardiovascular medicine. Applicants should have a strong technical background and experience with Python and deep learning. TV-L E13, fixed-term for 48 months.

The Department of Diagnostic and Interventional Radiology at Klinikum rechts der Isar, Technical University of Munich, is seeking a PhD Student (f/m/d) for multimodal embeddings and fusion architectures for digital patient twins in the EU research project TWIN-X.

Position details

• Position: PhD Student (f/m/d)

• Topic: Multimodal embeddings and fusion architectures for digital patient twins

• Project: TWIN-X, Digital Twins with Generative AI for Explainable Precision Medicine

• Institution: Department of Diagnostic and Interventional Radiology, Klinikum rechts der Isar, Technical University of Munich

• Employment: Full-time

• Salary: TV-L E13

• Duration: Fixed-term for 48 months

About TWIN-X

TWIN-X is a European Commission-funded research project under Horizon Europe. The consortium includes 18 partners from 12 European countries.

The project aims to develop digital patient twins by fusing multimodal clinical data into unified and explainable vector representations for clinical decision support in oncology and cardiovascular medicine.

Data modalities include:

• Radiology

• Pathology

• Genomics

• Laboratory values

• Clinical notes

The position is embedded in the core technical method development of TWIN-X and focuses on multimodal representation learning and the fusion of heterogeneous patient data.

Research focus

The PhD project may include the following areas:

• Multimodal representation learning Development of embedding models that map medical imaging, histopathology, genomics, laboratory data, and clinical text into a shared and explainable vector space.

• Fusion architectures Design and evaluation of modern fusion methods, such as cross-attention, mixture-of-experts models, and transformer-based multimodal models, for integrating heterogeneous and incomplete patient data.

• Self-supervised and contrastive learning Pre-training of multimodal foundation models on large clinical datasets.

• Explainability and robustness Development of methods to make learned representations interpretable, robust, and trustworthy for clinical use.

• Clinical validation Application and evaluation of the developed models in oncological and cardiovascular use cases together with clinical partners.

The exact focus will be determined together with the supervisor and can be tailored to the applicant’s interests and expertise.

Your responsibilities

• Independent research in multimodal representations and digital patient twins

• Development and implementation of deep learning methods for fusing heterogeneous clinical data

• Publication of research results at international conferences and in peer-reviewed journals

• Collaboration with clinicians, data scientists, and partners across the EU consortium

• Contribution to project deliverables and milestone reports

• Potential supervision of student theses

Your profile

• Master’s degree with very good to excellent grades in Computer Science, Medical Informatics, Physics, Mathematics, Biomedical Engineering, or a related field

• Strong programming skills in Python and experience with deep learning frameworks, preferably PyTorch

• Interest in multimodal AI, representation learning, medical imaging, or natural language processing

• Excellent English skills

• German is a plus but not required

• Self-motivation, teamwork skills, and enthusiasm for interdisciplinary research

Preferred qualifications

• Experience with multimodal models, vision-language models, or foundation models

• Experience with scientific writing and publishing

• Experience with medical datasets or clinical research

We offer

• Remuneration according to TV-L E13, full-time

• Fixed-term position for the project duration of 48 months

• PhD at TUM, one of Europe’s leading technical universities

• Access to extensive clinical datasets and high-performance computing infrastructure, including GPU clusters

• Integration into an international research network with 18 partners

• Opportunity to attend international conferences and workshops

• Flexible working hours and the option for remote work

Application

Please send your complete application documents by email.

Application documents should include:

• Cover letter

• CV

• Transcripts

• Publication list, if available

• Code portfolio or GitHub profile, if available

Contact

Prof. Dr. Lisa Adams;

PD Dr. med. Keno Bressem

Email: keno.bressem@tum.de

Equal opportunity statement

TUM is committed to increasing the proportion of women in its workforce. Applications from women are therefore expressly encouraged. Candidates with disabilities who are otherwise equally qualified will be given preference.

The position is suitable for disabled persons. Disabled applicants will be given preference in case of generally equivalent suitability, aptitude and professional performance.

Data Protection Information:

When you apply for a position with the Technical University of Munich (TUM), you are submitting personal information. With regard to personal information, please take note of the data protection information on collecting and processing personal data contained in your application in accordance with Art.13 of the General Data Protection Regulation (GDPR). By submitting your application, you confirm that you have acknowledged the above data protection information of TUM.

Kontakt: keno.bressem@tum.de