Design and optimization of a multi-module Compton camera system for thyroid scintigraphy
Year: 2026
Type: article
Abstract: Thyroid scintigraphy is a common clinical nuclear medicine procedure that provides both metabolic and structural information about the thyroid gland. Current clinical practice employs pinhole or parallel-hole collimators mounted on gamma cameras, which present several limitations including significant weight, potential safety hazards, and reduced photon detection efficiency. This work presents the initial simulation results towards a novel compact Compton camera system specifically designed for thyroid scintigraphy applications that eliminates the need for mechanical collimation. The proposed system utilizes a dual-layer detector configuration with GAGG:Ce scintillator crystals: a 2 mm thick scattering layer and a 10 mm thick absorbing layer. Monte Carlo simulations using GAMOS were employed to optimize the detector geometry and validate imaging performance. Two reconstruction algorithms were implemented and compared: Compton cone back-projection and list-mode maximum likelihood expectation maximization (LM-MLEM). MLEM reconstruction with 8 iterations achieves around 4 mm spatial resolution at a 5 cm source distance and can detect isolated sub-millimeter features. To acquire standard thyroid views (LAO, anterior, RAO) simultaneously and resolve depth ambiguity present in single-module Compton cameras, a three-module setup was tested, showing significant improvement in 3D reconstruction. This configuration could shorten examination time and eliminate the need for patient repositioning required by conventional methods. The simulation results provide validated design parameters for experimental prototype construction currently underway in our laboratory, where the predicted performance will be verified through experimental measurements.
Source: Physica Scripta
Authors: Asuman Kolbaşı, Cengizhan Öztürk, Bora Akgün
Language: English
Subfield: Radiation
Field: Physics and Astronomy
Domain: Physical Sciences
Open Access status: hybrid
Estimated APC paid: $2,400
OpenAlex ID: W7160500439