人工智能辅助的太赫兹成像无标记检测识别方法研究  

Research on Artificial Intelligence Assisted Terahertz Imaging Identification for Label-Free Detection

作　　者：王少娜 郭艳苗 郭泽康 李响 WANG Shaona;GUO Yanmiao;Guo Zekang;LI Xiang(School of Electronics and Information Engineering,Tiangong University,Tianjin 300387,China;Tianjin Key Laboratory of Optoelectronic Detection Technology and System,Tianjin 300387,China)

机构地区：[1]天津工业大学电子与信息工程学院,天津300387 [2]天津市光电检测技术与系统重点实验室,天津300387

出　　处：《测试技术学报》2024年第2期93-99,共7页Journal of Test and Measurement Technology

基　　金：国家自然科学基金资助项目(61901297)。

摘　　要：低强度激光治疗(Low-level Laser Therapy,LLLT)是一种可以提供局部治疗的新型无创光动力疗法。现有不同病症下光配方检测方法复杂而且对专业经验要求高,限制了LLLT的发展。为了检测急性肺损伤(Acute Lung Injury,ALI)的LLLT高效光配方,提出人工智能(Artificial Intelligence,AI)辅助的基于太赫兹(Terahertz,THz)成像的无标记识别方法。AI辅助的无标记THz成像可通过投票分类器自动识别。结果表明,AI辅助的THz成像自动识别方法对于不同光配方的LLLT其ALI的治疗效果识别率高达91.7%,可作为LLLT进一步开发和应用的新工具。Low-level laser therapy(LLLT)has been considered as a new noninvasive photodynamic therapy that provide local treatment.Currently,the development of LLLT is limited by the complexity of medical assay methods and the requirements of professional experience in identification of light formulas for different diseases.Here,a label-free identification method based on artificial intelligence(AI)assisted terahertz imaging is proposed for efficient light formulas in LLLT of acute lung injury(ALI).The AI-assisted label-free terahertz imaging is performed by automatic identification algorithm based on a voting classifier.The results indicate that the therapeutic effect of LLLT with different light wavelength and irradiation time for ALI can be identified by this method with a high accuracy of 91.7%,which may serve as a new tool for further development and application of LLLT.

关 键 词：低强度激光治疗 太赫兹成像 人工智能 随机森林 支持向量机 

分 类 号：Q631[生物学—生物物理学]