PhD defence Prodromos Gavriilidis

Keywords: Positron emission tomography, Gallium-68, Positron range correction, Deep learning

"Advanced Reconstruction and Correction Methods for Gallium-68 PE"

Positron emission tomography (PET) is an imaging technique widely used for cancer detection. PET relies on radiopharmaceuticals that contain radioisotopes. In recent years, new radiopharmaceuticals based on the radioisotope Gallium-68 have been emerged. However, Gallium-68 presents a challenge. It emits positrons that travel a relatively large distance before producing the annihilation photons detected by PET. This travelled distance is known as positron range and it can blur the PET images.

To correct for this effect, positron range correction techniques were developed. Additionally, deep learning models were also developed to enable lower administered activity or shorter PET acquisitions while simultaneously apply positron range correction.

The application of positron range correction enhanced the image quality and lesion detectability, especially for small size lesions in patient data. The deep learning models successfully enabled a reduction in the administered activity or shorter PET acquisitions while applying positron range correction, potentially reducing radiation exposure.

Click here for the live stream.