Study of imaging system in proton computed tomography with data acquisition from a monolithic active pixel sensor

Abstract

The first proton therapy system in Thailand has been installed at King Chulalongkorn Memorial Hospital (KCMH) since 2019. Apart from its clinical usage, Suranaree University of Technology (SUT) has initiated the collaborative research work with KCMH to explore a possibility of developing a proton computed tomography (pCT) prototype. Due to the proton’s depth-dose properties, this technique is more effective than photon treatment. Prior to proton therapy, pCT could help with the treatment planning. This technique simplifies proton treatment calculations since both processes involve the same particle interaction with matter. We simulated the experiment setup using G4beamline with the proton beam using a monolithic active pixel sensor (MAPS) as a proton tracker. The sensor’s output revealed that the ratio of data from protons interacting on the sensor was low, so we pre-processed the data with MATLAB by applying a mean filter to replace any empty pixels with the average of their nearby pixels. A 3D reconstruction was performed by stacking all the axial images reconstructed by the back projection method. The result shows that the material densities of reconstructed samples can be identified, however, it is still in the preliminary stage and not yet suitable for clinical trials. It is, however, possible to improve the quality of images and obtain a better 3D reconstruction for the pCT prototype by using a back-projection method.