Effect of adaptive statistical iterative reconstruction-V algorithm and deep learning image reconstruction algorithm on image quality and emphysema quantification in COPD patients under ultra-low-dose conditions.

This article investigates the impact of different CT image reconstruction algorithms—adaptive statistical iterative reconstruction-Veo (ASIR-V) and deep learning image reconstruction (DLIR)—on image quality and emphysema quantification in chronic obstructive pulmonary disease (COPD) patients using ultra-low-dose CT (ULDCT) scanning. In a prospective study of 62 COPD patients, ULDCT images reconstructed with medium-strength DLIR (DLIR-M) demonstrated comparable image noise, emphysema index (EI), and lung density metrics to standard-dose CT (SDCT) images reconstructed with filtered back projection (FBP), while reducing radiation dose by approximately 93.8%. DLIR-M also showed the strongest correlation between EI and pulmonary function test indices (FEV1/FVC and FEV1%), supporting its clinical feasibility for emphysema assessment under ultra-low-dose conditions. The study concludes that ULDCT combined with DLIR-M reconstruction provides image quality and quantitative emphysema analysis comparable to standard-dose CT, offering a promising approach to minimize radiation exposure in COPD evaluation and follow-up.