Development of a dosimeter prototype with machine learning based 3-D dose reconstruction capabilities.

Area of Science:

• Medical Physics
• Radiation Oncology
• Materials Science

Background:

• Modern radiation therapy requires precise verification of treatment plans and delivery.
• Existing 3-D dosimetry methods face limitations in accuracy and applicability.

Purpose of the Study:

• To develop and validate a novel water-equivalent solid 3-D dosimeter for radiation therapy verification.
• To assess the performance of a radiation-hard scintillating material in a prototype dosimeter.

Main Methods:

• Fabrication of a solid dosimeter using radiation-hard polysiloxane plastic doped with P-Terphenyl and BisMSB.
• Testing the prototype dosimeter with 6 MV and 10 MV x-ray beams.
• Utilizing a trained neural network for 3-D dose distribution reconstruction.

Main Results:

• The prototype dosimeter demonstrated water equivalency.
• Successful reconstruction of 3-D dose distributions was achieved.
• The novel scintillating material proved effective for dosimetry applications.

Conclusions:

• The developed solid 3-D dosimeter is a viable tool for radiation therapy verification.
• The novel radiation-hard material shows promise for advanced dosimetry.
• Machine learning effectively reconstructs 3-D dose distributions from the prototype detector.