Value of artificial intelligence-assisted diagnostic system for CT image interpretation in differential diagnosis of benign and malignant pulmonary nodules
10.13491/j.issn.1004-714X.2024.05.015
- VernacularTitle:CT图像人工智能辅助诊断系统鉴别肺结节良恶性的价值
- Author:
Xiaoqin SHEN
1
;
Hong LIANG
2
;
Xiaoqiong ZHU
1
Author Information
1. Department of Respiratory Medicine, Anyue County People’s Hospital, Ziyang 642350 China.
2. Department of Radiology, Anyue County People’s Hospital, Ziyang 642350 China.
- Publication Type:OriginalArticles
- Keywords:
Low-dose CT scan;
Artificial intelligence;
Pulmonary nodule;
Differential diagnosis;
Diagnostic value
- From:
Chinese Journal of Radiological Health
2024;33(5):578-583
- CountryChina
- Language:Chinese
-
Abstract:
Objective To compare artificial intelligence-assisted diagnostic system and conventional manual CT image interpretation for detection of positive pulmonary nodules and differential diagnosis of benign and malignant pulmonary nodules, and to provide a reference for the application of artificial intelligence in clinical screening for lung cancer. Methods Patients who underwent chest CT scans for pulmonary nodules from March 2019 to December 2023 were enrolled. The CT images were subjected to artificial intelligence-based and conventional manual CT image interpretation. The pathological examination results of pulmonary lesions served as a gold standard for comparison of artificial intelligence-based and conventional manual CT image interpretation in detection rate of positive pulmonary nodules and differential diagnosis of benign and malignant pulmonary nodules. Results A total of 327 positive pulmonary nodules were identified in 207 patients. The detection rate of positive pulmonary nodules was significantly higher with artificial intelligence-based CT image interpretation than with conventional manual CT image interpretation (95.72% vs. 86.85%; χ2=16.16, P < 0.01). Moreover, artificial intelligence-based CT image interpretation showed significantly higher detection rates for solid (χ2=7.71, P < 0.01) and ground-glass pulmonary nodules (χ2=5.80, P < 0.05) than conventional manual CT image interpretation. The detection rates for pulmonary nodules with < 1 cm (χ2=4.97, P < 0.05), 1 to < 2 cm (χ2=7.04, P < 0.01), and 2 to < 3 cm (χ2=4.91, P < 0.05) diameters were significantly higher with artificial intelligence-based CT image interpretation than with conventional manual CT image interpretation. The sensitivity, specificity, positive predictive value, negative predictive value, and accuracy for differential diagnosis of benign and malignant pulmonary nodules were 98.08%, 91.53%, 95.33%, 96.43%, and 95.71% with artificial intelligence-based CT image interpretation and 91.34%, 77.97%, 87.96%, 32.62%, and 86.50% with conventional CT image interpretation. The sensitivity (χ2=4.70, P < 0.05), specificity (χ2=4.20, P < 0.05), negative predictive value (χ2=65.28, P < 0.01), and accuracy (χ2=8.52, P < 0.01) were significantly higher with artificial intelligence-based CT image interpretation than with conventional manual CT image interpretation. However, there was no significant difference in the positive predictive value (χ2=3.80, P > 0.05). Conclusion Compared with conventional manual CT image interpretation, artificial intelligence-assisted diagnostic system for CT image interpretation can significantly increase the detection rate of positive pulmonary nodules and improve the efficiency of differential diagnosis of benign and malignant pulmonary nodules, so it deserves widespread applications in physical examination and early screening for lung cancer.
