Biomedical Physics

Course with Participations of Group Members

Offers for Theses in the Group

Convolutional neural networks and transfer learning for artefacts reduction in X-ray dark-field CT

Grating-based X-ray dark-field (DF) imaging uses scattering of X-rays to create an image of an object, rather than conventional X-ray attenuation. The combination of X-ray scattering with imaging allows us to map information about structures that are much smaller than the resolution of the imaging system over a large field of view. X-ray dark-field imaging can be combined with computed tomography (CT) to create three-dimensional images of the scattering distribution inside an object.  DF-CT was recently implemented for the first time into a clinical CT here at TUM

(https://www.bioengineering.tum.de/en/news/details/new-technology-for-clinical-ct-scans).

The goal of this project is to use convolutional neural networks (CNNs) to remove sampling artefacts in DF-CT images. Due to the unavailability of training data from the DF-CT machine, a technologically similar experimental setup and apply transfer learning will be used. The student will acquire, process and prepare training data, as well as train and apply CNNs.

Character of thesis work: experimental lab work/ data acquisition (50%) & computational/ image processing (50%)

Basic experience in image processing, CNNs, and/or Python programming are desirable.

For more information, please contact: Dr. Florian Schaff (florian.schaff@tum.de), or Prof. Franz Pfeiffer (franz.pfeiffer@tum.de).

suitable as

• Master’s Thesis Biomedical Engineering and Medical Physics

Supervisor: Franz Pfeiffer

Convolutional neural networks and transfer learning for artefacts reduction in X-ray dark-field CT

Grating-based X-ray dark-field (DF) imaging uses scattering of X-rays to create an image of an object, rather than conventional X-ray attenuation. The combination of X-ray scattering with imaging allows us to map information about structures that are much smaller than the resolution of the imaging system over a large field of view. X-ray dark-field imaging can be combined with computed tomography (CT) to create three-dimensional images of the scattering distribution inside an object.  DF-CT was recently implemented for the first time into a clinical CT here at TUM

(https://www.bioengineering.tum.de/en/news/details/new-technology-for-clinical-ct-scans).

The goal of this project is to use convolutional neural networks (CNNs) to remove sampling artefacts in DF-CT images. Due to the unavailability of training data from the DF-CT machine, a technologically similar experimental setup and apply transfer learning will be used. The student will acquire, process and prepare training data, as well as train and apply CNNs.

Character of thesis work: experimental lab work/ data acquisition (50%) & computational/ image processing (50%)

Basic experience in image processing, CNNs, and/or Python programming are desirable.

For more information, please contact: Dr. Florian Schaff (florian.schaff@tum.de), or Prof. Franz Pfeiffer (franz.pfeiffer@tum.de).

suitable as

• Master’s Thesis Applied and Engineering Physics

Supervisor: Franz Pfeiffer

Distortion correction for high-resolution quantitative X-ray imaging detectors

X-ray imaging detectors - in particular for high-resolution microscopy applications - may suffer from distortions, which degrade the image quality. This can have severe negative effects for quantitative applications, such as 3D micro-computed tomography. This project focuses on the characterisation of distortions of several X-ray imaging detectors at the Munich Compact Light Source, and the subsequent development of suitable correction methods.

Character of thesis work: mainly computational (image processing)  

For more information, please contact: Martin Dierolf (martin.dierolf@tum.de), Johannes Melcher (johannes.melcher@tum.de), or Franz Pfeiffer (franz.pfeiffer@tum.de)

suitable as

• Master’s Thesis Applied and Engineering Physics

Supervisor: Franz Pfeiffer

Distortion correction for high-resolution quantitative X-ray imaging detectors

X-ray imaging detectors - in particular for high-resolution microscopy applications - may suffer from distortions, which degrade the image quality. This can have severe negative effects for quantitative applications, such as 3D micro-computed tomography. This project focuses on the characterisation of distortions of several X-ray imaging detectors at the Munich Compact Light Source, and the subsequent development of suitable correction methods.

Character of thesis work: mainly computational (image processing)  

For more information, please contact: Martin Dierolf (martin.dierolf@tum.de), Johannes Melcher (johannes.melcher@tum.de), or Franz Pfeiffer (franz.pfeiffer@tum.de)

suitable as

• Master’s Thesis Biomedical Engineering and Medical Physics

Supervisor: Franz Pfeiffer

Estimation of Compton scattering in X-ray imaging using neural networks

Compton scattering is one of the two primary interactions of X-rays with matter in X-ray imaging (next to photoelectric absorption). In contrast to photoelectric absorption, Compton scattering is an inelastic scattering process during which X-ray photons are deflected and transfer some of their energy to the interaction partner, typically an electron. As a consequence, photons may still reach the detector after Compton interaction. In X-ray imaging, this leads to a smoothly varying background, which reduces contrast and is detrimental to quantitative imaging. Furthermore, the Compton scatter background is not uniform, but instead depends on the materials and their distribution within an imaged object. This makes a straight-forward analytical correction difficult, and existing tools to estimate the Compton background are limited in their accuracy and applicability.

 

The goal of this thesis is to develop methods to a) estimate the Compton scattering background from simple radiographs, and b) correct these images for it. This will be done using machine learning, in particular convolutional neural networks. The student will generate Monte-Carlo simulations based on the Geant4 platform, design and train neural networks, apply them for Compton scatter correction on clinical radiography images, and compare the results to existing approaches.

 

The project involves mostly data preparation, and computational work (85%, primarily Python, with potentially some C++ for Geant4), as well as experimental data collection (15%). The project will involve collaboration with the Radiology department at the TUM Hospital Klinikum rechts der Isar.

 

Basic experience in scientific programming, Monte-Carlo simulations, neural networks, and/or X-ray imaging are desirable.

 

For more information, please contact: Dr. Florian Schaff (florian.schaff@tum.de), or Prof. Franz Pfeiffer (franz.pfeiffer@tum.de).

suitable as

• Master’s Thesis Biomedical Engineering and Medical Physics

Supervisor: Franz Pfeiffer

Estimation of Compton scattering in X-ray imaging using neural networks

Compton scattering is one of the two primary interactions of X-rays with matter in X-ray imaging (next to photoelectric absorption). In contrast to photoelectric absorption, Compton scattering is an inelastic scattering process during which X-ray photons are deflected and transfer some of their energy to the interaction partner, typically an electron. As a consequence, photons may still reach the detector after Compton interaction. In X-ray imaging, this leads to a smoothly varying background, which reduces contrast and is detrimental to quantitative imaging. Furthermore, the Compton scatter background is not uniform, but instead depends on the materials and their distribution within an imaged object. This makes a straight-forward analytical correction difficult, and existing tools to estimate the Compton background are limited in their accuracy and applicability.

 

The goal of this thesis is to develop methods to a) estimate the Compton scattering background from simple radiographs, and b) correct these images for it. This will be done using machine learning, in particular convolutional neural networks. The student will generate Monte-Carlo simulations based on the Geant4 platform, design and train neural networks, apply them for Compton scatter correction on clinical radiography images, and compare the results to existing approaches.

 

The project involves mostly data preparation, and computational work (85%, primarily Python, with potentially some C++ for Geant4), as well as experimental data collection (15%). The project will involve collaboration with the Radiology department at the TUM Hospital Klinikum rechts der Isar.

 

Basic experience in scientific programming, Monte-Carlo simulations, neural networks, and/or X-ray imaging are desirable.

 

For more information, please contact: Dr. Florian Schaff (florian.schaff@tum.de), or Prof. Franz Pfeiffer (franz.pfeiffer@tum.de).

suitable as

• Master’s Thesis Applied and Engineering Physics

Supervisor: Franz Pfeiffer

High-sensitivity grating-based phase-contrast imaging at the Munich Compact Light Source - Computational Part

Using phase-contrast as alternative imaging contrast for X-rays can considerably improve the imaging results for biomedical specimens. This project will focus on the development of an algorithmic framework for a high-sensitivity and high-resolution grating-based phase-contrast micro-tomography setup at the Munich Compact Light Source for investigating soft-tissue biomedical samples, such biopsies. 

Character of thesis work: mainly computational (image processing/ reconstruction)  

For more information, please contact: Martin Dierolf (martin.dierolf@tum.de), Johannes Brantl (johannes.brantl@tum.de), or Franz Pfeiffer (franz.pfeiffer@tum.de)

suitable as

• Master’s Thesis Biomedical Engineering and Medical Physics

Supervisor: Franz Pfeiffer

High-sensitivity grating-based phase-contrast imaging at the Munich Compact Light Source - Computational Part

Using phase-contrast as alternative imaging contrast for X-rays can considerably improve the imaging results for biomedical specimens. This project will focus on the development of an algorithmic framework for a high-sensitivity and high-resolution grating-based phase-contrast micro-tomography setup at the Munich Compact Light Source for investigating soft-tissue biomedical samples, such biopsies. 

Character of thesis work: mainly computational (image processing/ reconstruction)  

For more information, please contact: Martin Dierolf (martin.dierolf@tum.de), Johannes Brantl (johannes.brantl@tum.de), or Franz Pfeiffer (franz.pfeiffer@tum.de)

suitable as

• Master’s Thesis Applied and Engineering Physics

Supervisor: Franz Pfeiffer

High-sensitivity grating-based phase-contrast imaging at the Munich Compact Light Source - Experimental Part

Using phase-contrast as alternative imaging contrast for X-rays can considerably improve the imaging results for biomedical specimens. This project will focus on the experimental construction of a high-sensitivity and high-resolution grating-based phase-contrast micro-tomography setup at the Munich Compact Light Source for investigating soft-tissue biomedical samples, such as biopsies. Character of thesis work: mainly experimental For more information, please contact: Martin Dierolf (martin.dierolf@tum.de), Johannes Brantl (johannes.brantl@tum.de), or Franz Pfeiffer (franz.pfeiffer@tum.de)

suitable as

• Master’s Thesis Applied and Engineering Physics

Supervisor: Franz Pfeiffer

High-sensitivity grating-based phase-contrast imaging at the Munich Compact Light Source - Experimental Part

Using phase-contrast as alternative imaging contrast for X-rays can considerably improve the imaging results for biomedical specimens. This project will focus on the experimental construction of a high-sensitivity and high-resolution grating-based phase-contrast micro-tomography setup at the Munich Compact Light Source for investigating soft-tissue biomedical samples, such as biopsies. Character of thesis work: mainly experimental For more information, please contact: Martin Dierolf (martin.dierolf@tum.de), Johannes Brantl (johannes.brantl@tum.de), or Franz Pfeiffer (franz.pfeiffer@tum.de)

suitable as

• Master’s Thesis Biomedical Engineering and Medical Physics

Supervisor: Franz Pfeiffer

Implementation of a grating-based interferometer for X-ray vector radiography at the Munich Compact Light Source

Grating-based X-ray dark-field (DF) imaging uses scattering of X-rays to create an image of an object, rather than conventional X-ray attenuation. The combination of X-ray scattering with imaging allows us to map information about structures that are much smaller than the resolution of the imaging system over a large field of view. The fact that the used gratings typically are one-dimensional can be leveraged to obtain an orientation dependent dark-field signal in a technique called X-ray vector radiography (XVR). Applications of XVR include the determination of the fibre orientation in reinforced composite materials, or characterization of the anisotropic structure in trabecular bones.

The goal of this project is to implement an experimental XVR setup at the Munich Compact Light Source (MuCLS - https://www.bioengineering.tum.de/en/central-building/munich-compact-light-source). The student will help with the design, implementation, and characterization of the X-ray grating interferometer setup, and conduct their own XVR experiments.

Character of thesis work: experimental lab work/ controls/ data acquisition (50%) & computational/ simulation/image processing (50%)

Basic experience in X-ray imaging, and/or Python programming are desirable.

For more information, please contact: Dr. Florian Schaff (florian.schaff@tum.de), or Prof. Franz Pfeiffer (franz.pfeiffer@tum.de).

suitable as

• Master’s Thesis Biomedical Engineering and Medical Physics

Supervisor: Franz Pfeiffer

Implementation of a grating-based interferometer for X-ray vector radiography at the Munich Compact Light Source

Grating-based X-ray dark-field (DF) imaging uses scattering of X-rays to create an image of an object, rather than conventional X-ray attenuation. The combination of X-ray scattering with imaging allows us to map information about structures that are much smaller than the resolution of the imaging system over a large field of view. The fact that the used gratings typically are one-dimensional can be leveraged to obtain an orientation dependent dark-field signal in a technique called X-ray vector radiography (XVR). Applications of XVR include the determination of the fibre orientation in reinforced composite materials, or characterization of the anisotropic structure in trabecular bones.

The goal of this project is to implement an experimental XVR setup at the Munich Compact Light Source (MuCLS - https://www.bioengineering.tum.de/en/central-building/munich-compact-light-source). The student will help with the design, implementation, and characterization of the X-ray grating interferometer setup, and conduct their own XVR experiments.

Character of thesis work: experimental lab work/ controls/ data acquisition (50%) & computational/ simulation/image processing (50%)

Basic experience in X-ray imaging, and/or Python programming are desirable.

For more information, please contact: Dr. Florian Schaff (florian.schaff@tum.de), or Prof. Franz Pfeiffer (franz.pfeiffer@tum.de).

suitable as

• Master’s Thesis Applied and Engineering Physics

Supervisor: Franz Pfeiffer

Radioluminescence microscopy for organoid specimens

Radioluminescence microscopy (RLM) is a novel approach for high-resolution imaging of the radionuclide uptake in living cells, particularly in organoid systems. This project will focus on the development of an experimental setup, which allows imaging the radionuclide distribution with a few micro-meter resolution, using a scintillator-lens CCD system. This project will be carried out in collaboration with the department of nuclear medicine at the TUM university hospital Klinikum rechts der Isar.

Character of thesis work: experimental (50%) / computational (50%)

For more information, please contact: Martin Dierolf (martin.dierolf@tum.de), Franz Pfeiffer (franz.pfeiffer@tum.de)

suitable as

• Master’s Thesis Applied and Engineering Physics

Supervisor: Franz Pfeiffer

Radioluminescence microscopy for organoid specimens

Radioluminescence microscopy (RLM) is a novel approach for high-resolution imaging of the radionuclide uptake in living cells, particularly in organoid systems. This project will focus on the development of an experimental setup, which allows imaging the radionuclide distribution with a few micro-meter resolution, using a scintillator-lens CCD system. This project will be carried out in collaboration with the department of nuclear medicine at the TUM university hospital Klinikum rechts der Isar.

Character of thesis work: experimental (50%) / computational (50%)

For more information, please contact: Martin Dierolf (martin.dierolf@tum.de), Franz Pfeiffer (franz.pfeiffer@tum.de)

suitable as

• Master’s Thesis Biomedical Engineering and Medical Physics

Supervisor: Franz Pfeiffer

Spectral material decomposition with dual-energy, photon-counting X-ray detectors for industrial applications

Despite being meanwhile well established and broadly available in medical imaging applications, spectral material decomposition using photon-counting hybrid pixel detectors is presently still underused in industrial imaging tasks. The main goal of this project is therefore to translate the existing theoretical and experimental knowledge from photon-counting biomedical imaging applications to industrial inspection applications. The work includes numerical programming tasks, such as implementing and refining decomposition algorithms,  and experimental tasks, such as taking several best practice measurement to classify different material separation applications for industrial end-users.

This master thesis will be carried out in collaboration with an external industrial collaborator located in the Munich area.

Character of thesis work: experimental physics (50%) & image processing (50%)

For more information, please contact: Franz Pfeiffer (franz.pfeiffer@tum.de)

suitable as

• Master’s Thesis Applied and Engineering Physics

Supervisor: Franz Pfeiffer

Spectral material decomposition with dual-energy, photon-counting X-ray detectors for industrial applications

Despite being meanwhile well established and broadly available in medical imaging applications, spectral material decomposition using photon-counting hybrid pixel detectors is presently still underused in industrial imaging tasks. The main goal of this project is therefore to translate the existing theoretical and experimental knowledge from photon-counting biomedical imaging applications to industrial inspection applications. The work includes numerical programming tasks, such as implementing and refining decomposition algorithms,  and experimental tasks, such as taking several best practice measurement to classify different material separation applications for industrial end-users.

This master thesis will be carried out in collaboration with an external industrial collaborator located in the Munich area.

Character of thesis work: experimental physics (50%) & image processing (50%)

For more information, please contact: Franz Pfeiffer (franz.pfeiffer@tum.de)

suitable as

• Master’s Thesis Applied and Engineering Physics

Supervisor: Franz Pfeiffer

Current and Finished Theses in the Group