Cao Yuqi, Huang Pingjie, Chen Jiani, Ge Weiting, Hou Dibo, Zhang Guangxin
State Key Laboratory of Industrial Control Technology, College of Control Science and Engineering, Zhejiang University, Hangzhou, China.
Cancer Institute (Key Laboratory of Cancer Prevention and Intervention, China National Ministry of Education), the Second Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou 310009, Zhejiang, China.
Biomed Opt Express. 2020 Jan 21;11(2):982-993. doi: 10.1364/BOE.381660. eCollection 2020 Feb 1.
Terahertz technology has been widely used as a nondestructive and effective detection method. Herein, terahertz time-domain spectroscopy was used to detect drug-induced liver injury in mice. Firstly, the boxplots were used to detect abnormal data. Then the maximal information coefficient method was used to search for the features strongly correlated with the degree of injury. After that, the liver injury model was built using the random forests method in machine learning. The results show that this method can effectively identify the degree of liver injury and thus provide an auxiliary diagnostic method for detecting minor liver injury.
太赫兹技术已被广泛用作一种无损且有效的检测方法。在此,太赫兹时域光谱被用于检测小鼠药物性肝损伤。首先,使用箱线图检测异常数据。然后使用最大信息系数法搜索与损伤程度高度相关的特征。之后,利用机器学习中的随机森林法建立肝损伤模型。结果表明,该方法能够有效识别肝损伤程度,从而为检测轻度肝损伤提供一种辅助诊断方法。