Master Thesis: Deep Learning for Sarcoma Diagnostics

21.02.2025, Abschlussarbeiten, Bachelor- und Masterarbeiten

Abstract:

Sarcomas are rare and heterogeneous tumors that require advanced diagnostic tools for early detection, classification, and treatment planning. Deep learning methodologies offer powerful solutions to analyze medical imaging data, aiding in automated diagnostics and decision support. This master thesis opportunity explores novel deep learning techniques, including self-supervised learning, generative modeling, and multi-modal learning, to enhance the detection, segmentation, and classification of bone tumors. Students will have the opportunity to contribute to cutting-edge research in medical AI and work with real-world imaging datasets, including MRI, CT, X-ray scans, and possible medical reports.

Topic:

This thesis offering includes multiple topics related to deep learning for sarcoma diagnostics. Depending on the student’s interests and background, the research can be tailored to focus on several different areas, for example:

• Self-supervised contrastive learning for multi-modal bone tumor representation.
• Unsupervised tumor detection in X-ray and MRI bone tumors using generative models.
• Weakly-supervised learning for bone tumor classification and localization.
• Cross-modal knowledge transfer for bone tumor detection in low-resource imaging modalities.

Prerequisites:

• Advanced knowledge of deep learning.
• Beneficial but not necessary: experience in medicine/oncology.
• Preferred starting date: we offer master thesis opportunities year-round, depending on our availability. Interested candidates are encouraged to contact us.

What we offer:

• Very rare medical data with high potential for publication.
• Highly educated & interdisciplinary environment.
• Top-level hardware for scientific computing.
• Constant feedback from medical and computer science experts.

How to apply:

Send an email to anna.curto-vilalta@tum.de with the following:

• Your CV
• Academic transcripts
• A brief introduction about yourself and your motivation
• A short text outlining your interests within the mentioned research areas, specifying what aspects you would be most interested in exploring.

Kontakt: anna.curto-vilalta@tum.de