Evidence Hub

Comparison of DLIR-H and ASIR algorithms for image reconstruction in low-dose chest CT of pediatric mycoplasma pneumoniae pneumonia: a cross-sectional study

Y. Zou, Hai Bi, Shouhong Yang, Rui-shi Li, J. Tan, Qingwei Guo, Xiudi Lu, Ji-Hong Liu

Translational Pediatrics, 2026

235 participants

Abstract

Background Pediatric Mycoplasma pneumoniae pneumonia (MPP) accounts for a significant number of cases of community-acquired pneumonia, and chest computed tomography (CT) is critical for its early diagnosis and prognosis assessment. However, clinical utilization is restricted due to concerns regarding radiation exposure. Adaptive statistical iterative reconstruction (ASIR) has been extensively employed in low-dose pediatric CT but suffers from noise texture distortion. In contrast, deep learning image reconstruction (DLIR) has shown promise for enhancing image quality in adult CT. Nevertheless, the comparative effectiveness of DLIR and ASIR in low-dose chest CT for pediatric MPP remains unvalidated, presenting a significant knowledge gap that this study endeavours to bridge. Methods In this retrospective cross-sectional study, consecutive pediatric patients who underwent low-dose chest CT (September 2023–March 2024) who were positive for MP-IgM and negative for other respiratory pathogen antibodies were included; patients with incomplete data, poor scan cooperation or a body mass index (BMI) ≥25 kg/m2 were excluded. The raw CT data were reconstructed via ASIR at 0%, 50%, and 80% blending factors and DLIR at low, medium, and high-strength (DLIR-H) levels, using a standard kernel and a 1.25 mm slice thickness/interval. The standard deviation value of subcutaneous fat, the signal-to-noise ratio (SNR) of the lung parenchyma, and the contrast-to-noise ratio (CNR) of MPP were used…

How strong is the evidence?

Evidence Hub reads this as a cross-sectional. Its evidence level is 4 of 5 (Low · observational). Observational designs can show an association but not prove cause. Reported size: 235 participants or records.

How the five levels of evidence work. The level is assigned automatically from the abstract and can be wrong, so check the paper.

Read the paper

Related studies

Go deeper in Evidence Hub

Evidence Hub adds a study data sheet (design, sample size, evidence level, key numbers), an AI summary and a chat grounded in the paper's text.

Open this study Sign up free