Dortmund 2021 – wissenschaftliches Programm

Bereiche | Tage | Auswahl | Suche | Aktualisierungen | Downloads | Hilfe

ST: Fachverband Strahlen- und Medizinphysik

ST 3: Artificial Intelligence in Medical Physics

ST 3.3: Vortrag

Dienstag, 16. März 2021, 14:30–14:45, STa

A neural network for event identification in a Compton camera for imaging prompt gamma rays in proton therapy — •Pascal Wirtz, Ronja Hetzel, Jonas Kasper, and Achim Stahl — RWTH Aachen University - Physics Institute III B, Aachen, Germany

The utilization of proton beams for cancer treatments has gained popularity because it allows a more precise deposition of energy due to the Bragg peak. One big challenge is the live monitoring of the beam. The SiFi-CC project develops a Compton camera based on scintillating fibres which detects the prompt gamma rays in order to determine the penetration depth of the beam. A neural network is developed to reconstruct the direction of the prompt gamma rays. This network consists of three parts. One classifies whether the detected data represents a Compton event, while the other parts identify the locations and energies of the electron and photon of the Compton event. The classification of the event type already works quite well. Therefore the aim is to improve the reconstruction of the energies and locations of the particles in the Compton event in order to get better in determining the direction of the prompt gamma ray. With the directions of those an image reconstruction can be done to identify whether the position of the Bragg peak is inside the tumor.