Speaker
Description
Currently, cancer is one of the most frequent death causes in the world and
radiation therapy is used in approximately 50% of patients diagnosed with
cancer. This implies the need of the treatment to be as efficient and safe
as possible. In this work, a novel reconfigurable Dose-3D detector intended for
a full spatial therapeutic dose reconstruction to improve radiotherapy
treatment planning by providing a breakthrough detector with active voxels
is presented. The device is comprising a customizable detector head, a
scalable data acquisition system (including hardware, firmware and low-level
software) and a state of the art high-level software.
The detector head is being designed as a set of 3D-printed scintillator
pieces, whose shape and arrangement can be changed to accommodate patient's
needs. A feasibility study was done to assure the quality of the detector
manufactured using the aforementioned method. The results show, that the
light output of 3D-printed scintillators provides sufficient signal to noise
ratio for the project.
The data acquisition system (DAQ) is designed to accommodate the changing
geometry by varying the number of slices, each capable of aggregating
64 detection channels into 1 Gbps Ethernet link. The low-level software can
interact with virtually any number of DAQ units. Prototype devices have been
tested successfully with the whole detection chain in place.
The high-level software is being designed to automatically convert medical
data (CT scans) into accurate 3D models of the tumor and neighbouring cells
using machine learning. Obtained geometry will be used to create dedicated
detector head for the patient, as well as an environment for dose simulation
in GEANT.
In conclusion, the research undertaken until now confirm the possibility
to build a device to greatly personalise and improve radiotherapy planning
and effectiveness.
