Journal article
Development of a dosimeter prototype with machine learning based 3-D dose reconstruction capabilities
Biomedical physics & engineering express, Vol.8(1), p.15009
01/01/2022
DOI: 10.1088/2057-1976/ac396c
PMID: 34768242
Abstract
A 3-D dosimeter fills the need for treatment plan and delivery verification required by every modern radiation-therapy method used today. This report summarizes a proof-of-concept study to develop a water-equivalent solid 3-D dosimeter that is based on novel radiation-hard scintillating material. The active material of the prototype dosimeter is a blend of radiation-hard peroxide-cured polysiloxane plastic doped with scintillating agent P-Terphenyl and wavelength-shifter BisMSB. The prototype detector was tested with 6 MV and 10 MV x-ray beams at Ohio State University's Comprehensive Cancer Center. A 3-D dose distribution was successfully reconstructed by a neural network specifically trained for this prototype. This report summarizes the material production procedure, the material's water equivalency investigation, the design of the prototype dosimeter and its beam tests, as well as the details of the utilized machine learning approach and the reconstructed 3-D dose distributions.
Details
- Title: Subtitle
- Development of a dosimeter prototype with machine learning based 3-D dose reconstruction capabilities
- Creators
- G M Finneman - Coe CollegeO H Eichhorn - Coe CollegeN R Meskell - Coe CollegeT W Caplice - Coe CollegeA D Benson - Coe CollegeA S Abu-Halawa - Coe CollegeG L Ademoski - Coe CollegeA C Clark - Coe CollegeD S Gayer - Coe CollegeK N Hendrickson - Coe CollegeP A Debbins - University of IowaY Onel - University of IowaA S Ayan - The Ohio State UniversityU Akgun - Coe College
- Resource Type
- Journal article
- Publication Details
- Biomedical physics & engineering express, Vol.8(1), p.15009
- DOI
- 10.1088/2057-1976/ac396c
- PMID
- 34768242
- ISSN
- 2057-1976
- eISSN
- 2057-1976
- Grant note
- name: NSF RUI, award: 1746230; name: NSF REU, award: 1659581
- Language
- English
- Date published
- 01/01/2022
- Academic Unit
- Physics and Astronomy
- Record Identifier
- 9984428665202771
Metrics
8 Record Views