Wave based image reconstruction in Ultrasound Tomography

Research topic/area

Medical Imaging / 3D Ultrasound Tomography

Type of thesis

Bachelor / Master

Start time

-

Application deadline

31.10.2026

Duration of the thesis

6 Months (adaptable)

Description

At the Institute for Data Processing and Electronics (IPE) a new imaging method for early breast cancer detection, named “3D Ultrasound Computer Tomography” is currently developed. For this ultrasound based imaging method the breast is embedded in the water bath and surround by approx. 2000 ultrasound transducers. The raw data is acquired by sending out an approx. spherical wave into the measurement aperture such that the ultrasound interacts with the tissue resulting in reflection, attenuation, refraction etc. The resulting signals, so called A-Scans representing the pressure over time, are acquired by all ultrasound transducers around the breast. The entire acquisition process results in millions of such A-Scans which are afterwards used to reconstruct 2D or 3D images involving signal and image processing algorithms.

For transmission imaging ray based methods are applied since they can be computed efficiently in reasonable time for clinical diagnosis. However the approximation of the ultrasound propagation along (straight) rays is a coarse assumption which leads to a considerable limitation of the resolution. Properly modeling the wave physics would enable to increase the resolution by an order of magnitude, however at the cost of an dramatic increase in computational complexity.

This works aims to explore a first prototypic implementation of a Full Waveform Inversion (FWI) based on an open source toolbox, which has previously been used to reconstruct 2D slices from the 3D Ultrasound Tomography data. These promising results should be extended to allow the reconstruction of arbitrary planes in the 3D setup with adaptable slice thickness and orientation. With this setup a first reconstructions of experimental data should be carried out and the influence of the selection of the subset of the aperture on the convergence and image quality should be analyzed. Furthermore a study based on simulated data should be carried out to find a good tradeoff between data sparsity, image quality and data acquisition time.

Requirement

Requirements for students

• Preferably basic knowledge in MATLAB programming, medical imaging, ultrasound

Faculty departments

• Engineering sciences
  Electrical engineering & information technologies
  Informatics
  Computer Science
  Electrical Engineering and Information Technology
  Medical technology
• Natural sciences and Technology
  Physics

Supervision

Title, first name, last name

Dr. Torsten Hopp

Organizational unit

IPE

Email address

torsten.hopp@kit.edu

Link to personal homepage/personal page

Website

Application via email

Application documents

• Cover letter
• Curriculum vitae
• Grade transcript

E-Mail Address for application
Senden Sie die oben genannten Bewerbungsunterlagen bitte per Mail an torsten.hopp@kit.edu


Back