Work

Association of AI-Derived Quantitative CT Parameters of Airway, Emphysema, and Pulmonary Vasculature with Lung Cancer: A Cross-Sectional Analysis

Year: 2026

Type: article

Abstract: BACKGROUND: Given multiple risk factors for lung cancer, this study explored associations between lung cancer and AI-derived quantitative chest computed tomography (CT) parameters of emphysema, airways, and pulmonary vasculature. METHODS: This retrospective single-center study (December 2020-February 2023) analyzed relevant parameters of the left upper lobe (LUL) and right upper lobe (RUL) in 170 lung cancer patients and 126 healthy individuals. Subgroups were defined by cancer-free lobes (129 patients/126 controls for LUL; 120 patients/126 controls for RUL). Univariate and multivariate binary logistic regression analyses were used for analysis. RESULTS: The emphysema-related 15th percentile of CT attenuation values (PI-15) was significantly associated with lung cancer, with lower values in patients’ LUL. Pulmonary vascular parameters (diameter, count, area at 6 mm/24 mm from the lung surface) differed significantly; the patients had smaller diameters, higher counts, and larger areas at 6 mm in the LUL. Airway parameters (Awt-Pi10, level 6 wall thickness) were higher in patients’ LUL. Multivariate regression identified PI-15 and vascular diameters (6 mm/24 mm) in LUL [area under the curve (AUC) = 0.841, 95% confidence interval (95% CI): 0.789–0.892] and vascular diameters (6 mm/24 mm) and vascular count at 24 mm from the lung surface in RUL (AUC=0.819, 95% CI:0.766–0.872) as significant predictors (all P<0.001). CONCLUSION: AI-derived quantitative CT parameters of emphysema, vasculature, and airways are associated with lung cancer and may serve as complementary tools for clinical risk assessment.

Source: Current Medical Imaging Formerly Current Medical Imaging Reviews

DOI: 10.2174/0115734056413168251203110120

Authors: Xiaojun Zhou, Qi Dai, Wei Lu, Zizhen Yang, Jianjun Zheng, Jingfeng Zhang

Institution: Ningbo No. 2 Hospital

Language: English

Topic: Lung Cancer Diagnosis and Treatment

Subfield: Pulmonary and Respiratory Medicine

Field: Medicine

Domain: Health Sciences

SDG: Good health and well-being

Open Access status: hybrid

Funder: Medical Science and Technology Project of Zhejiang Province