Theses

Masters and Bachelors theses and research internships

If you are interested in a Masters or Bachelor thesis project or a research internship (Forschungspraxis) in our group we are happy to propose a concrete problem related to our current interests. Our group focuses on machine learning and optimization, deep learning for inverse problems, and DNA data storage and DNA information technologies. To get an idea about our current research, please check out our recent papers at google scholar. Projects usually involve a mixture of theory and applied work and require strong interest documented by excellent grades in relevant subjects such as linear algebra, probability and statistics, machine learning, signal processing, optimization, or related courses.

Below is a list of open topics. If you are interested in one of the projects, please send an email directly to the superviser and cc reinhard.heckel@tum.de. Include your transcript of records from TUM and your curriculum vitae and the planned start and end dates.

The list of topics is sometimes incomplete, and we are also happy to propose other topics if there is a good fit. If you are interested in a topic related to our current interests that is not listed below, please reach out to reinhard.heckel@tum.de and again include your transcript of records and CV.

External projects: If you are planning to carry our our project externally, for example at a company or another university, and you want us to supervise the project, please send an email to reinhard.heckel@tum.de including the name and contact of the exernal supervisor, your transcipt of records and CV, and a project description and an explanation how the project is related to our expertise. We can only supervise external projects if they are related to our expertise and current research interests.

Open projects

Robust fine-tuning of deep learning models for accelerated MRI reconstruction

We’re looking for a Master’s student to explore robust fine-tuning strategies for pre-trained deep learning models for accelerated MRI. While fine-tuning improves performance on the target dataset, it sacrifices robustness to out-of-distribution data. This project aims to better understand fine-tuning for image reconstruction and in particular accelerated MRI with the goal of proposing robust fine-tuning methods for accelerated MRI.
Supervisor: Kang Lin (ka.lin@tum.de)
Prerequisites: Strong coding skills with PyTorch. A good understanding of deep learning frameworks.
Type of project: Masters thesis or Reserach Internship (Forschungspraxis)

 

Deep learning based fetal magnetocardiography system

We are looking for a Master’s student to work on deep learning based reconstruction for a fetal magnetocardiography system, in a interdisciplinary collaboration with Prof. Fierlinger and Prof. Wacker-Gussmann. The goal of the thesis is to develop deep learning based signal reconstruction techniques for this new medical imaging and sensing technology.
Supervisor: Reinhard Heckel, Kang Lin
Prerequisites: Strong coding skills, specially in pytorch. A good understanding of deep learning for inverse problems.
Type of project: Masters thesis


Learning a full camera pipeline

A camera’s image processing pipeline usually consists of many different image processing modules, e.g. debayering, denoising, filtering. Many of these steps rely on measurements that enable some parameter settings, that are used in a part of the image pipeline. Successful learning-based approaches like the approach of “Learning to see in the dark” present a different approach to obtain a high-quality image by learning the full camera pipeline. While the work in this paper uses a convolutional network, transformer-based approaches including an attention mechanism are have shown greater success on image processing tasks recently. This student project investigates the possibility of learning a full camera pipeline for real camera data based on fusion of classical and learning based approaches to find the optimal tradeoff between state-of-the-art image quality and efficiency on a real camera hardware.
Supervisor: This project is supervised by Dr. Seybold at ARRI and carried out at ARRI, the supervision from the TUM side is by Reinhard Heckel.
Prerequisites: Coding skills, interest in imaging and signal processing.
Type of project: Details of the scope can be adjusted to the kind of thesis (Bachelor thesis, Master thesis, research internship)

 

Self-supervised reconstruction of PET Scans

We are looking for a Master's student to work on reconstructing PET scans using self-supervised deep learning techniques. Since obtaining ground truth PET images is difficult and expensive, self-supervised training has emerged as a useful tool. In other medical image domains such as CT and MRI self-supervised techniques have proven to be successful. Our aim is to test those techniques on PET imaging. To learn more about the project, please contact Youssef Mansour.
Supervisor: Youssef Mansour (y.mansour@tum.de)
Prerequisites: Strong coding skills, specially in pytorch. A good understanding of deep learning for inverse problems.
Type of project: Masters thesis

 

A Test-Time-Training Approach for Image Denoising under Distirbution Shifts

We are looking for a Master's student to work on a project of improving a method that finetunes a trained model at test time to adapt to different distribution shifts for image denoising. The different shifts include natural and medical images, synthetic noise, and real-world camera and microscope noise. Your task is to understand and reproduce the method as as well adding ablation studies and extra comparisons with other baselines. You will also contribute to improving the performance of the method, and making it more intuitive. To learn more about the project, please contact Youssef Mansour.
Supervisor: Youssef Mansour (y.mansour@tum.de)
Prerequisites: Strong coding skills, specially in pytorch. A good understanding of deep learning frameworks. Interest in collaborative work.
Type of project: Masters thesis or Reserach Internship (Forschungspraxis)

 

Estimating the age of a DNA sample from sequencing data 

We are looking for a Master’s student to develop, implement, and test an idea to estimate the age of a given DNA sample from sequencing data alone. First you would perform literature research on existing approaches and then implement and test a few concrete ideas in python.
Supervisor: Reinhard Heckel (reinhard.heckel@tum.de)
Prerequisites: Interest in interdisciplinary work and data analysis.
Type of project: Masters thesis

 

Cardiac magnetic resonance imaging with neural networks 

We’re looking for a Master’s student to work on a new deep neural network based method for imaging a non-static object, specifically a beating heart. You would implement an idea for a neural network based method in pytorch, evaluate the method on real data within a collaboration with the University Hospital, and there is also room for developing your own ideas.
Supervisor: Reinhard Heckel (reinhard.heckel@tum.de)
Prerequisites: Ideally knowledge of deep learning for inverse problems and python and pytorch programming experience.
Type of project: Masters thesis

 

Current and past theses in the group

Serdar Caglar, ``Test Time Training for denoising distribution shifts’’, Project/Forschungspraxis, ongoing

Mamdouh Aljoud, ``Filtering techniques in next generation multimodal datasets’’, Master's thesis, 2023

Cheng Yan, ``Meta-Learning for mulit-task MRI reconstruction’’, Master's thesis, 2023

Francesco Bollero, ``Data pruning for image reconstruction’’, Master's thesis, 2023

Faidra Patsatzi, ``Randomized smoothing for inverse problems’’, Bachelor's thesis, 2023

Xiaodong Lei, ``Adversarial robustness of deblurring methods’’, Project/Forschungspraxis, 2023

Dogukan Atik, ``Scaling laws for self-supervised image denoising’’, Master's thesis, 2023

Mamdouh Aljoud, ``Deep networks for nanopore basecalling’’, Project/Forschungspraxis, 2023

Rafael Vorländer, ``Reimplementing CryoGAN’’, Project/Forschungspraxis, 2023

Juan Cao, ``Uncertainty quantification methods for compressed sensing’’, Project/Forschungspraxis, 2023

Litao Li, ``Datawork for accelerated MRI’’, Master's thesis, 2023

Guang Chai, ``Evaluating deep-learning based imaging systems’’, Master's thesis, 2023

Johannes Kunz, ``Generative models for Cardiac Magnetic Resonance Imaging’’, Master's thesis, 2023

Ali Can, ``Understanding the Contribution of Training Samples on a Prediction of a Single Test Image in a Denoising Task Using Attention Mechanism’’, Project/Forschungspraxis, 2022

Deniz Uysal,  ``Spectral Computed Tomography Image Reconstruction’’, Master's thesis, 2022

Xuyang Zhong, ``Self-Supervised Learning for Image Denoising’’, Master's thesis, 2022

Johannes Kunz, ``Dynamic MRI reconstruction’’, Project/Forschungspraxis, 2022

Weixing Wang, ``Graph neural networks for clustering and aligning DNA sequences for DNA storage'', Project/Forschungspraxis, 2022

Yundi Zhang, ``Coordinate-based image priors'', Master's thesis, 2022.

Samuel Eadie, ``Rate-Distortion Stochastic Autoencoding for Robust Representation Learning and Out-of-Distribution Detection'' (carried out at Bosch Research), Master's thesis, 2022

Kang Lin, ``Transformers for image recovery'', Master's thesis, 2021.

Frederik Fraaz, ``Image recovery with invertible neural networks'', Master's thesis, 2021.

Youssef Mansour, ``Neural network architectures for image recovery and denoising'', Master's thesis, 2021.

Benedikt Böck, ``Multiplicative filter networks for image processing applications'', Project/Forschungspraxis, 2021

Mohamed Ketata,  ``Data standardisation, multi-domain learning, and artifact robustness for improved MRI'', Bachelor's thesis, 2021.

Deniz Uysal, ``A simple encoder and decoder for DNA data storage with Polar codes'', Project/Forschungspraxis, 2021.

Yundi Zhang, ``Deep matrix decoder for collaborative filtering'', Project/Forschungspraxis, 2021.

Youssef Mansour, ``Ensembles of image reconstruction method for MRI'', Project/Forschungspraxis, 2021.

Jacob Geussen, ``Diffusion MRI denoising with neural networks'', Bachelor's thesis, 2020.

Lena Heidemann, ``FastMRI with untrained neural networks'', Master's thesis, 2020.

Tobit Klug, ``Image separation with untrained neural networks'', Master's thesis, 2020.

Oleksii Khakhlyuk, ``Convolutional neural networks with fixed kernels'', Bachelor's thesis, 2019. 

Zi Yang, ``Probabilistic matching networks for few-shot learning'', Master's thesis, 2019.