Calibration of Gamma Ray Impacts in Monolithic-Based Detectors Using Voronoi Diagrams
Molecular imaging systems, such as positron emission tomography (PET), use detectors providing energy and a 3-D interaction position of a gamma ray within a scintillation block. Monolithic crystals are becoming an alternative to crystal arrays in PET. However, calibration processes are required to c...
| Autores: | , , , , |
|---|---|
| Tipo de recurso: | artículo |
| Estado: | Versión publicada |
| Fecha de publicación: | 2020 |
| País: | España |
| Institución: | Consejo Superior de Investigaciones Científicas (CSIC) |
| Repositorio: | DIGITAL.CSIC. Repositorio Institucional del CSIC |
| OAI Identifier: | oai:digital.csic.es:10261/420044 |
| Acceso en línea: | http://hdl.handle.net/10261/420044 https://api.elsevier.com/content/abstract/scopus_id/85085993104 |
| Access Level: | acceso abierto |
| Palabra clave: | Calibration processes Gamma ray detectors Monolithic crystals Positron emission tomography (PET) SiPM Voronoi diagrams |
| Sumario: | Molecular imaging systems, such as positron emission tomography (PET), use detectors providing energy and a 3-D interaction position of a gamma ray within a scintillation block. Monolithic crystals are becoming an alternative to crystal arrays in PET. However, calibration processes are required to correct for nonuniformities, mainly produced by the truncation of the scintillation light distribution at the edges. We propose a calibration method based on the Voronoi diagrams. We have used 50 × 50 × 15 mm3 LYSO blocks coupled to a 12 × 12 SiPMs array. We have first studied two different interpolation algorithms: 1) weighted average method (WAM) and 2) natural neighbor (NN). We have compared them with an existing calibration based on 1-D monomials. Here, the crystal was laterally black painted and a retroreflector (RR) layer added to the entrance face. The NN exhibited the best results in terms of XY impact position, depth of Interaction, and energy, allowing us to calibrate the whole scintillation volume. Later, the NN interpolation has been tested against different crystal surface treatments, allowing always to correct edge effects. Best energy resolutions were observed when using the reflective layers (12%- 14%). However, better linearity was observed with the treatments using black paint. In particular, we obtained the best overall performance when lateral black paint is combined with the RR. |
|---|