Deep Learning Reconstruction Enables Substantial Radiation Dose Reduction in Ultra-High-Resolution Temporal Bone CT: a Body Donor Study.
Authors
Affiliations (5)
Affiliations (5)
- Department of Neuroradiology, University Medical Center of the Johannes Gutenberg University Mainz, Mainz, Germany.
- Institute of Anatomy, University Medical Center of the Johannes Gutenberg University Mainz, Mainz, Germany.
- Department of Integrated Basic Medical Science, Macroscopic Anatomy, Medical University Lausitz - Carl Thiem, Cottbus, Germany.
- Department of Neuroradiology, University Medical Center of the Johannes Gutenberg University Mainz, Mainz, Germany. [email protected].
- Department of Diagnostic and Interventional Neuroradiology, Hannover Medical School, Hannover, Germany. [email protected].
Abstract
Deep learning reconstruction may enable substantial radiation dose reduction in ultra-high-resolution (UHR) temporal bone CT, but its performance across clinically relevant dose levels remains insufficiently defined. This study evaluated vendor-specific deep learning reconstruction (DLR) compared with hybrid iterative reconstruction (HIR) in UHR volume-mode temporal bone CT and aimed to propose an indication-adapted framework for dose optimization. In this single-center body donor study, 20 temporal bones from 10 donors without pathologic findings were scanned at 11 dose levels (CTDI<sub>vol</sub>, 5.1-30.7 mGy). Images were reconstructed using HIR at standard resolution (0.5 mm, 512<sup>2</sup> matrix) and UHR (0.25 mm, 1024<sup>2</sup> and 2048<sup>2</sup> matrices), as well as DLR (AiCE Inner Ear) at UHR (0.25 mm, 1024<sup>2</sup> matrix). Three radiologists independently assessed 10 anatomical structures using a 5-point Likert scale. Signal-to-noise ratio was measured in predefined regions. Exploratory structure-specific generalized estimating equation models were used to estimate the probability of achieving optimal image quality as a function of radiation dose and reconstruction technique. DLR significantly improved visualization of osseous structures compared with HIR (p < 0.001). The exploratory dose-response analysis demonstrated reconstruction-specific trajectories, with DLR achieving higher predicted probabilities of optimal image quality than HIR at lower dose levels. For routine clinical indications, diagnostically adequate image quality was achieved with DLR at a CTDI<sub>vol</sub> of 13.3 mGy, representing a 48% dose reduction compared with HIR (CTDI<sub>vol</sub> 25.6 mGy). For high-detail assessment of delicate microanatomical structures, including the stapes and cochlea, DLR achieved excellent visualization at a CTDI<sub>vol</sub> of 17.4 mGy, corresponding to a 43% dose reduction compared with HIR (CTDI<sub>vol</sub> 30.7 mGy). Vendor-specific DLR enables substantial radiation dose reduction in UHR volume-mode temporal bone CT while maintaining high diagnostic image quality. These findings support an indication-adapted framework for dose optimization, suggesting a routine clinical protocol at CTDI<sub>vol</sub> 13.3 mGy and a microanatomy-optimized protocol at CTDI<sub>vol</sub> 17.4 mGy.