期刊文献+

基于中医声诊数字化指标的肺结节辨识研究

Research on pulmonary nodule based on digital auscultation indicators of traditional Chinese medicine sound diagnosis
原文传递
导出
摘要 目的:探索肺系疾病特别是肺结节的声诊标志,为通过声诊语音特征辨识肺结节的早期诊断方式提供客观依据。方法:于2022年11月—2023年12月北京中医药大学第三附属医院采集198例肺系疾病患者[有肺结节组(PN组)103例,无肺结节组(NPN组)95例],以及100名健康受试者[健康对照组(HC组)]的语音信号。利用中医声诊数字化指标提取算法将语音信号转化为数字化指标,比较各组语音特征,并构建肺结节的机器学习分类模型。结果:在423项声诊指标中,58项在3组间差异有统计学意义(P<0.05),其中28项指标仅可以区分出肺系疾病,19项指标可以特异性区分出肺结节,其中6项在3组中均有显著性差异(P<0.05)。支持向量机(SVM)模型对肺结节的识别精确率为0.91。结论:中医声诊数字化指标能有效区分肺系疾病患者与健康人群,特别是识别肺结节,为肺结节的早期诊断提供了新视角。SVM模型在肺结节识别中表现良好,也为中医声诊客观化以及智能临床应用提供了新的可能。 Objective:To investigate the acoustic diagnostic markers for pulmonary diseases,particularly pulmonary nodules,and provide an objective basis for early diagnosis through acoustic feature recognition.Methods:From November 2022 to December 2023,voice signals were collected from 198 patients with pulmonary diseases at the Third Affiliated Hospital of Beijing University of Chinese Medicine,including 103 with pulmonary nodules(PN group)and 95 without(NPN group),as well as 100 healthy controls(HC group).Using a digital acoustic diagnosis algorithm,voice signals were converted into quantitative indicators.The acoustic features of each group were compared,and a machine learning classification model for pulmonary nodules was constructed.Results:Among 423 acoustic diagnostic indicators,58 showed statistically significant differences across the three groups(P<0.05).Of these,28 indicators distinguished pulmonary diseases,19 specifically identified pulmonary nodules,and 6 showed significant differences in all three groups(P<0.05).The support vector machine(SVM)model achieved an accuracy of 0.91 in identifying pulmonary nodules.Conclusion:Digital acoustic diagnostic indicators effectively distinguish between pulmonary disease patients and healthy individuals,particularly in identifying pulmonary nodules.This approach provides a new perspective for early diagnosis,and the SVM model demonstrates strong performance in nodule detection,offering new potential for objective and intelligent clinical applications of acoustic diagnosis in traditional Chinese medicine.
作者 张晶新 李腾腾 李艺博 朱琦 芦煜 卢涛 ZHANG Jingxin;LI Tengteng;LI Yibo;ZHU Qi;LU Yu;LU Tao(School of Life Sciences,Beijing University of Chinese Medicine,Beijing 102401,China;Institute of Information on Traditional Chinese Medicine,China Academy of Chinese Medical Sciences,Beijing 100700,China)
出处 《中华中医药杂志》 CAS CSCD 北大核心 2024年第11期6144-6147,共4页 China Journal of Traditional Chinese Medicine and Pharmacy
基金 国家自然科学基金项目(No.82104739)。
关键词 肺结节 中医声诊 数字化 智能诊断 支持向量机模型 Pulmonary nodules Traditional Chinese medicine sound diagnosis Digitalization Intelligent diagnosis Support vector machine(SVM)model
  • 相关文献

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部