Seminaria

Dnia 2023-11-28 o godzinie 14:15 w Sali 2011 Wydziału Fizyki UwB, mgr Konrad Skórkiewicz z Zakładu Fizyki Medycznej, Instytutu Fizyki, Uniwersytetu Jagiellońskiego w Krakowie wygłosi wykład pt:

Obrazowanie PET/CT w diagnostyce guzów neuroendokrynnych – co może jeszcze zrobić fizyk medyczny?

Serdecznie zapraszamy

Andrzej Maziewski

Jerzy Przeszowski

Obrazowanie PET/CT w diagnostyce guzów neuroendokrynnych – co może jeszcze zrobić fizyk medyczny?

PET/CT imaging in the diagnosis of neuroendocrine tumors – what else can a medical physicist do?

Konrad Skórkiewicz

Faculty of Physics, Astronomy and Applied Computer Science Jagiellonian University in Kraków

Abstract

PET/CT diagnostics uses iterative image reconstruction techniques to obtain reliable information about the biodistribution of radiotracers in the human body. There are usually several image reconstruction algorithms available in clinically used devices. Changing the image reconstruction algorithm or modifying it by changing the number of iterations and subsets affects the results of patient examinations.

The conducted scientific research included an analysis of available image reconstruction algorithms in a clinically used PET/CT device installed at the University Hospital in Krakow (Discovery MI DR, GE). The aim of the analysis was to select an algorithm intended to reconstruct examinations of oncological patients treated and diagnosed at the Department of Nuclear Medicine. In particular, the aim of the conducted research was to select an algorithm that allows for more accurate diagnosis of patients with neuroendocrine tumors, in whom tests with 68Ga isotope-labelled somatostatin analogues are performed at various stages of the disease.

The research was divided into two stages. In the first stage, the PET/CT examination of the NEMA phantom was analyzed. The study analyzed the effect of changing the β parameter (range 150 – 950, step 100) in the VP FX+Q.Clear and VP HD+Q.Clear algorithms on the semi-quantitative Standard Volume Uptake (SUV) scores and the ratio of the measured sphere to the background (CR coefficient). The homogeneity of the background in the body part of the phantom was also assessed with Minkowski analysis.

In the second part of the work, a comparative analysis of three image reconstruction algorithms VP FX+Q.Clear, VP HD+Q.Clear and VP FX was performed using the results of patients with advanced forms of cancer. The group of patients was carefully selected, because the impact of the reconstruction algorithm changes was tested, especially for small (< 2cm) metastases of neuroendocrine tumours.

The results obtained allowed us to assess the effect of the beta parameter value on Q.Clear reconstruction and the suitability of using Q.Clear reconstruction at various stages of cancer, especially at its early stage. The selected value of the β parameter is in the range of 350-450. It was found that it is more advantageous to use the VP FX+Q.Clear algorithm, which uses information from the correction of attenuation and time of flight of photons generated during the annihilation phenomenon.

Key words: Q.Clear algorithm, neuroendocrine tumours, NEMA standards

Disclosure

The research was carried out with the financial support of two research grants:

• N17/MNW/000016, N17/MNS/000060,
• “Research support module” as a part of the “Excellence Initiative – Research University” program at the Jagiellonian University.