中国中西医结合影像学杂志2025，Vol.23Issue(6)：717-720,4.DOI:10.3969/j.issn.1672-0512.2025.06.009
肺结节患者中医证型与人工智能风险程度分析的相关性研究
Study on correlation between risk levels analyzed by artificial intelligence and traditional Chinese medicine syndrome in patients with pulmonary nodules
摘要
Abstract
Objective:To observe and analyze the correlation between traditional Chinese medicine(TCM)syndrome and the artificial intelligence(AI)risk levels in patients with pulmonary nodules.Methods:A total of 141 patients with pulmonary nodules were collected.AI models were used to analyze chest CT images,and the patients were classified into five TCM syndrome types,pattern of lung-spleen qi deficiency,lung yin deficiency pattern,pattern of blood stasis due to qi deficiency,phlegm heat accumulating in the lung pattern,qi stagnation damaging the lung pattern.Afterwards,the correlation between TCM syndrome and the risk level predicted by AI was evaluated.Results:A total of 1 198 nodules were detected in the 141 patients,with 4 cases being solitary nodules and 137 cases being multiple nodules.According to the quartile of the number of nodules,141 patients were divided into 4 groups,Group A with 1-3 nodules,Group B with 4-6 nodules,Group C with 7-11 nodules,and Group D with more than 11 nodules.Except for Group A,the proportion of high-risk was higher than that of medium-risk and low-risk in other groups.Among the 141 patients,the most common TCM syndrome type was pattern of blood stasis due to qi deficiency,accounting for 41.84%,followed by pattern of lung-spleen qi deficiency(24.11%),lung yin deficiency pattern(16.31%),qi stagnation damaging the lung pattern(12.06%)and phlegm heat accumulating in the lung pattern(5.68%).There were differences in the risk level of different TCM syndromes(χ2=50.38,P<0.001).In pattern of blood stasis due to qi deficiency,pattern of lung-spleen qi deficiency and phlegm heat accumulating in the lung pattern,the proportion of high-risk was significantly higher than that of medium-risk and low-risk(all P<0.05),while in pattern of lung-spleen qi deficiency,lung yin deficiency pattern and qi stagnation damaging the lung pattern,the proportion of medium-risk was significantly lower than that of high-risk and low-risk(all P<0.05).Conclusions:There is a certain correlation between TCM syndrome and the AI analysis of pulmonary nodule risk level,and different treatments can be carried out according to different syndromes in clinical practice.关键词
肺结节/人工智能/中医证型/体层摄影术,X线计算机Key words
Pulmonary nodules/Artificial intelligence/Traditional Chinese medicine syndrome/Tomography,X-ray computed引用本文复制引用
周海伦,孙玲玲,许玲,龚亚斌,刘睿潮,阿格尔,征宗梅,杜怡曌,王梦琪,杨文笑,王芹,焦丽静..肺结节患者中医证型与人工智能风险程度分析的相关性研究[J].中国中西医结合影像学杂志,2025,23(6):717-720,4.基金项目
国家重点研发计划项目(2022YFC3500203) （2022YFC3500203）
上海市炎癌转化病证生物学前沿科学研究基地(2021KJ03-12). （2021KJ03-12）
