Index

Automated Leptomeningeal Collateral Scoring in Acute Ischemic Stroke Using Deep Learning

Implant Object Detection in Intraoperative X-Ray Images

Thesis Start:
October 2025 or later

Your Profile and Skills:

• Successful completion of courses from our lab: (Advanced) Deep Learning / Pattern Recognition / Pattern Analysis
• Proficiency in Python programming and experience with PyTorch
• Fundamental knowledge about medical imaging and image processing
• Strong analytical, structured, and quality-oriented working style
• Ability to work independently while enjoying a collaborative team environment
• Strong communication skills in English

Application:
Please send your transcript of records, CV, and a small motivation letter on why you would be interested in the topic only to joshua.scheuplein@fau.de
Note: Applications not following these requirements will not be considered!

Uncertainty Estimation on Semantic Segmentation for Microscopy Data

In microscopy, many common data analysis tasks rely on an initial semantic segmentation step. Microscopy data are very diverse, and thus this segmentation might fail due to being out-of-distribution (OOD). For users to know whether the downstream tasks are possible or accurate, it is necessary to assess the accuracy of the semantic segmentation step. This can be done through uncertainty estimation of the predictions, either at the image or pixel level. To address this, we are conducting detailed research focusing on uncertainty estimation methods across four key categories: Deterministic, Bayesian Neural Networks (BNN), Ensemble, and Test Time Augmentation (TTA). This work aims to explore both well-established and emerging methods for uncertainty estimation in semantic segmentation applied to microscopy data.

Benchmarking State-of-the-Art Transformers for Handwritten Document Layout Analysis

Vision-Language Models in Radiology

1. Extend existing dataset with synthetically generated data
2. Train multimodal vision-language model
3. Perform extensive evaluation of the model on public datasets:
   • Investigate and apply suitable evaluation metrics.
   • Research state-of-the-art methods for comparison.
4. (Optional: Contribute to writing a research paper based on the results.)

1. Experience with PyTorch.
2. Experience with training deep learning models.
3. Ability to attend in-person meetings.

1. Curriculum Vitae (CV).
2. Short motivation letter (max. one page).
3. Transcript of records.

Send your application with the subject “Application VLM Radiology + your full name” to Lukas.Buess@fau.de.

Enhancing Financial QA with Hybrid Retrieval and Semantic Tagging

From Pixels to Structure: Analysis of Lightweight Vision-Language Models for Document OCR and Structured Output Generation

Investigation on Object Detection in Industrial Settings Centered on Extended Reality Platforms Through Generation and Utilization of Synthetic Data from CAD Models

Vision Language Models for Patient Retrieval in Radiation Therapy

Federated Learning for Medical Vision-Language Models

This thesis investigates how federated learning can be applied to train vision-language models in the medical domain while preserving patient privacy. The work focuses on enabling multi-institutional collaboration without sharing sensitive data, supporting the development of secure and scalable AI solutions for healthcare.