Answers are grounded in this paper and cite their sources.
Traditional back-projection reconstruction requires a high number of projections, whereas iterative reconstruction techniques deliver substantially better results in limited-projection scenarios. Consequently, their ability to reduce radiation exposure has attracted significant attention. However, these methods typically require a precise tuning of several hyperparameters, which can have a major impact on reconstruction quality. Manually setting these parameters is time-consuming and increases the workload for human operators. In this paper, a fully automatic parameter optimization framework is introduced that can be applied to different Cone-beam computed tomography (CBCT) iterative reconstruction algorithms to determine optimal parameters without requiring reference reconstruction. The proposed method incorporates a modified crow search algorithm (CSA) featuring a superior set–dependent local search mechanism, a search-space-aware global search strategy, and an objective-driven balance between local and global search. Additionally, to ensure an effective initial population, a chaotic diagonal linear uniform initialization scheme is proposed that accelerates algorithm convergence. The performance of the proposed framework was evaluated on four imaging machines and six real datasets, as well as three different iterative reconstruction methods with the highest number of tunable parameters. The proposed method outperformed manual settings and CSA, with 3.87% improvement in average fitness and 4.25% and 3.85% improvements on benchmark no-reference learning-based quality metrics. In addition, qualitative results indicate better visual preservation of the proposed method by maintaining fine details sharply. The overall performance on nine benchmark functions against eleven bio-inspired metaheuristics, demonstrates the proposed framework effectiveness and robustness across all cases.
No takes yet. Share an insight, caveat, or question.
MohammadiNasab et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: