Basic description of the topic:
Ultrasound imaging plays an important role in modern diagnostic and interventional procedures; however, the interpretation of ultrasound images and the accurate spatial localization of the examined structures remain challenging tasks. Phantom studies performed in a simulated environment provide a safe and reproducible way to investigate ultrasound imaging, external camera-based tracking, and image-based object localization. Such systems can provide a basis for the later development of navigation, robotic, augmented reality-based or AI-supported interventional solutions.
The candidate’s task is to develop a data acquisition and evaluation system for ultrasound phantom studies in the simulated operating room environment of the Óbuda University MedTech Innovation and Education Center. The work includes the examination of phantoms containing hidden structures, the acquisition of ultrasound image data, and the external camera-based tracking of the position or motion of the ultrasound transducer. The task may include the integration of ultrasound images and external camera data, the detection or segmentation of hidden objects in ultrasound images, and the evaluation of their spatial localization accuracy. During the development, 3D Slicer-based solutions may also be used for the visualization, synchronization or processing of ultrasound and tracking data.
Detailed tasks:
- Literature overview in the field of ultrasound imaging, tracked ultrasound systems, medical image navigation, phantom studies and ultrasound image segmentation;
- Getting acquainted with the operating room unit, ultrasound devices, camera-based tracking possibilities and relevant measurement infrastructure of the MedTech Innovation and Education Center;
- Development of an ultrasound phantom environment containing hidden structures or objects, or definition of an examination protocol for existing phantoms;
- Acquisition of ultrasound image and video data under different examination settings, transducer positions and phantom configurations;
- External camera-based tracking of the position or motion of the transducer, and investigation of possible methods for integrating ultrasound and external camera data;
- Getting acquainted with 3D Slicer, SlicerIGT, PLUS Toolkit or OpenIGTLink-based tools and applying them, if necessary, for data visualization, synchronization or processing;
- Development of an image processing or machine learning-based method suitable for the detection, segmentation or localization of hidden objects visible in ultrasound images;
- Evaluation of localization or segmentation accuracy based on known phantom geometry or reference markers;
- Testing the developed system in a simulated operating room environment and evaluating the results.
Over the course of the project, the student will have the opportunity to become familiar with the research infrastructure of the Óbuda University MedTech Innovation and Education Center and the Antal Bejczy Center for Intelligent Robotics, and to get involved in the research and development activities carried out there.