CT signs combined with quantitative indexes in the differential diagnosis of mycoplasma pneumoniae pneumonia in adults
10.3969/j.issn.1002-1671.2025.08.008
- VernacularTitle:CT征象联合定量指标在成人肺炎支原体肺炎鉴别诊断中的应用
- Author:
Dian LING
1
;
Xiang LI
;
Yu LI
;
Shuang GENG
Author Information
1. 江汉大学医学部,湖北 武汉 430056;华中科技大学同济医学院附属武汉市中心医院影像诊断科,湖北 武汉 430014
- Publication Type:Journal Article
- Keywords:
artificial intelligence;
mycoplasma pneumoniae pneumonia;
bacterial pneumonia;
high-resolution computed tomography
- From:
Journal of Practical Radiology
2025;41(8):1289-1293
- CountryChina
- Language:Chinese
-
Abstract:
Objective To evaluate the value of high-resolution computed tomography(HRCT)signs combined with artificial intelligence(AI)quantitative indexes in differentiating mycoplasma pneumoniae pneumonia(MPP)from bacterial pneumonia(BP)in adults.Methods A total of 100 patients with adult MPP and 100 with adult BP were selected.Univariate and multivariate logistic regression analyses were used to identify independent factors that contribute to the differentiation of these two types of pneumonia.The receiver operating characteristic(ROC)curve and area under the curve(AUC)were employed to evaluate the discrimination ability of the two types of pneumonia.Results A comparison of gender and age between the two groups revealed statistically significant differences(P＜0.05).A comparison of bronchial wall thickening,bud sign,and solid shadows between the two groups yielded statistically significant differences(P＜0.05),while small nodules did not(P＞0.05).AI-based quantification revealed significant differences in distribution of inflammatory foci,consolidation/total lesion ratio,ground-glass opacity(GGO)/total lesion ratio,consolidation density,and GGO density(P＜0.05).Univariate and multivariate logistic regression analyses revealed that left upper lobe(％),left lower lobe(％),right upper lobe(％),right middle lobe(％),and right lower lobe(％)were independent factors that contributed to the identification of the two types of pneumonia.The AUC for HRCT signs alone,AI quantitative indexes alone,and HRCT signs combined with AI quantitative indexes in differentiating the two types of pneumonia were 0.715[95％confidence interval(CI)0.643-0.787],0.822(95％CI 0.764-0.881),and 0.853(95％CI 0.801-0.905),respectively.The utilization of HRCT signs combined with AI quantitative indexes had been demonstrated to possess a substantial capacity to accurately identify two types of pneumonia.Conclusion HRCT signs combined with AI quantitative indexes can improve the accuracy of differentiating MPP from BP in adults.
