We are offering theses on a number of topics, including medical robotics, mobile robotics, navigation and image guidance, medical image analysis, machine learning, treatment planning and optimization.

Below, you will find summaries of our research interests. If you are interested in a topic and there is no specific thesis available, feel free to contact us. We have a number of further topics evolving with our daily research, and we are open to interesting topics beyond this.

When contacting us, please attach your transcript of records, both from bachelor and master. Also, briefly explain your technical interests and skills (e.g. programming, machine learning, robotics, image processing).

 nils.gesserttuhh.de     +49 (0)40 42878 3083      E.3389

Deep Learning for Medical Image Analysis

Deep learning methods are today's standard approach for medical image analysis. Typical learning problems are disease detection or tumor and organ segmentation. Deep learning models such as Convolutional Neural Networks (CNNs) or Recurrent Neural Networks (RNNs) directly learn the relevant properties from image source.

For example, the imaging modality intravascular optical coherence tomography (IVOCT) is typically used to detect plaque deposits in coronary arteries in order to prevent coronary heart disease. Currently, medical practitioners manually examine the IVOCT images which is time consuming and subjective. This motivates automated image analysis with deep learning methods.

Current thesis topics:

  • Semi-Supervised Deep Learning with IVOCT Image Data for Plaque Detection [studIP Link]

Deep Learning for Force Estimation and Tracking

Typically, deep learning and machine learning methods are used for the analysis of patient images for disease related problems. However, they can also be applied to tracking and force estimation problems that arise during surgery. For example, an intraoperative imaging modality in conjunction with a machine learning model can be used to track surgical tools and patients or to estimate forces that tools are exerting on patient tissue.

Current thesis topics:

  • Surgical Tool Tracking and Accurate Tool Control Using Multi-Input-Multi-Output Machine Learning methods [studIP Link]


The general requirements are:

  • Very good academic record
  • Very good programming skills (e.g. Python/Java/C/C++)
  • Background knowledge on machine learning and deep learning
  • Motivation and the ability to think critically and work independently

Please provide in your E-Mail:

  • Bachelor certificate and current transcript of records
  • Curriculum Vitae
  • A short text where you highlight your technical interests and related prior experiences

 mareike.wendebourgtuhh.de      +49 (0)40 42878 3547       E.3087

Tissue Phantoms

At MTEC, we often utilize gelatin phantoms as a substitute for human tissue. The process of gelatin phantom production is prone to variation, causing differences in the mechanical properties of the produced phantoms. Accurate insights into the factors influencing the phantom's mechanical properties may help to improve reproducability of phantoms with similar mechanical properties.

Seizure Detection

Nocturnal tonic-clonic seizures can be life-threatening when undetected. The use of depth cameras may facilitate a comfortable and affordable seizure detection system for clonic seizures. A possible thesis topic involves the further development of an existing seizure detection method, acquisition of clinical data and evaluation of the method's seizure detection capabilities.

 martin.gromniaktuhh.de      +49 (0)40 42878 3828       E.3088

Image Guidance and Navigation

In Image-guided medical procedures, surgeons use tracked instruments together with real-time images and potentially preoperative images to guide the procedures. Images can be acquired with different imaging modalities. At MTEC our focus is ultrasound, optical coherence tomography, and 3D cameras.

Current thesis topics:

  • Development of a Data Aquisiton Setup for EEG Electrode Detection Opens external link in new window[studIP Link]
  • Deep Learning-Based Needle Pose Estimation from 3D Camera Data Opens external link in new window[studIP Link]
  • Ultrasound to 3D Camera Calibration for Needle Guidance Opens external link in new window[studIP Link]

Haptic Feedback

In the context of robot-assisted minimally invasive surgery, the use of haptic feedback force has the potential to greatly improve the dexterity of the surgeon. At MTEC we are examining how general-purpose light-weight robot (LWRs) can be used as high quality haptic feedback devices.

Current thesis topics:

  • Design of a Force Feedback Controller for an LWR robot with the goal of steering a surgery device

Object Tracking

In a current research project, the correlation between nutrition, age and mobility should be investigated. For this purpose, zebrafishes will be used as model organisms. The individual fishes should be tracked based on camera image and a metric for their mobility will be calculated based on their trajectories. Multiple cameras facing different sides of the aquarium will be used to minimize occlusions between fishes.

Current thesis topics:

 maximilian.neidhardttuhh.de       +49 (0)40 42878 4639      E.1051


Optical Coherence Tomography is an optical imaging modality based on interferometric measurement of reflections of near-infrared light from different tissue depths. It offers a field of a view few millimeters but micrometer-scale spatial resolution.

The very high temporal resolution of modern OCT systems make it ideal for imaging shear waves which travel at high speed through tissue. The spreading of the wave can be directly linked to the elastic properties of a potential malignant tissue.

Current thesis topics:

Ultrasound Imaging

Medical ultrasound has been long established in the clinical routine and is still today advancing. Our lab offers two research ultrasound machines which can be used in a variety of scenarios such as:

  • Tracking approaches with a robot
  • Online image segmentation
  • Shear wave elastography imaging.
  • Image processing (beamforming)

Please contact me if you are interested in one of the above topics. In general you should bring enthusiasm for experimental lab work, good programming skills (C++/Python/Matlab) and the ability to work independently.

lennart.bargsten@tuhh.de       +49 (0)40 42878 3547       E.3087

Deep Learning and Other Methods Applied at Intravascular Ultrasound Image Data

Intravascular ultrasound (IVUS) is a medical image acquisition modality, where a catheter with an attached ultrasound probe is used to generate images from the inside of blood vessels (particularly coronary arteries). It can be used to detect stenoses or other vascular diseases, as well as for inspecting stents.

Deep learning tasks include the segmentaion of certain layers of the artery wall, detection of stenoses or stents or generating new labeled images be means of generative adversarial networks (GANs). GANs can also be used to enhance the image quality, to reduce the intensity of artifacts or to determine distributions of echogenicity or tissue density.

Current thesis topics:

  • Application of GANs for augmentation of limited IVUS data
  • Segmentation of IVUS pullbacks by means of recurrent neural networks
  • Extraction of 3D stent geometry from IVUS pullbacks
  • Transform disordered IVUS pullbacks into 3D artery representations

 schlaefertuhh.de       +49 (0)40 42878 3050      E.3085

Generally, I supervise all theses offered through our institute and I will be actively involved in theses projects. For more information, please see the topics offered through the research assistants.

However, there are number of toy projects I foster myself. Often these are longer shot problems that may have some impact in the future. For example, if you are interested in mobile robotics, please contact me. Also, I am happy to discuss your ideas if they are related to our research.