Application value of AI-assisted diagnosis combined with low-dose thin-section CT in the detection of pulmonary nodules and the interpretation of benign and malignant conditions
Xia Yuanjian
Chen Fucheng
Liu Wei
Chen Wei
Abstract：Objective To analyze the application effect of AI-assisted diagnosis combined with low-dose CT thin-layer scanning in the detection of pulmonary nodules and the interpretation of benign and malignant conditions.Methods A total of 75 patients with pulmonary nodules admitted to Songyuan Hospital of Integrated Traditional Chinese and Western Medicine from April 2024 to April 2025 were selected and underwent low-dose CT thin-layer CT scanning or AI-assisted diagnosis combined with low-dose CT thin-layer scanning.Taking clinicopathological examination as the gold standard,the influence of simple low-dose CT thin-layer scanning and the use of AI-assisted diagnosis schemes on the diagnostic efficacy of benign and malignant pulmonary nodules were analyzed.Results Clinicopathological examination revealed that among the 75 patients with pulmonary nodules,41 were benign and 34 were malignant.Low-dose CT thin-layer scanning detected 33 benign cases and 27 malignant cases.AI-assisted diagnosis combined with low-dose CT thin-layer scanning detected 39 benign cases and 32 malignant cases.The results of Kappa consistency analysis showed that the kappa value of AI-assisted diagnosis combined with low-dose CT thin-layer scan was 0.774,which was higher than the 0.609 of low-dose CT thin-layer scan(P<0.05).The sensitivity of AI-assisted diagnosis combined with low-dose CT thin-layer scanning was 95.12%and the specificity was 94.12%,which were higher than those of low-dose CT thin-layer scanning.There was no significant difference in the data comparison(χ2=2.847,P=0.092;χ2=2.049,P=0.152).The accuracy of AI-assisted diagnosis combined with low-dose CT thin-layer scanning was 94.67%,which was higher than that of low-dose CT thin-layer scanning,with a significant difference(χ2=6.027,P=0.014).Conclusion Compared with simple low-dose CT thin-layer scanning,the use of AI-assisted diagnosis can further improve the diagnostic efficiency of benign and malignant pulmonary nodules.
Keywords：Pulmonary nodulesArtificial intelligenceLow doseCT thin-layer scan
Publication Date：2025-05-20
Online Publishing Date：2025-11-20(First online date of this platform, not the publication date of the document)
Pages：4( 63-66 )