提出了两种有效混合物成分的定性与定量分析算法。针对混合物中每种成分太赫兹光谱已知的情况,提出了已知混合物成分光谱分析法,该方法具有步骤简单、样本需求量少的优点。而针对成分过多或部分成分光谱无法有效获得的情况,提出了未知混合物成分光谱分析法,该方法适用于多成分混合物,具有精度较高的优点,样本需求量多于已知混合物成分光谱分析法,但计算量相比传统算法大幅缩减。相关工作为生物医学领域中关键物质的识别和测量提供了新的思路和方法。

This study proposes two effective algorithms for the qualitative and quantitative analysis of components in a mixture. For cases wherein the spectrum of each component in the mixture can be obtained in advance, we develop a simple process, namely “known component analysis (KCA),” which requires a small number of samples. In contrast, for cases wherein a mixture contains many components or the spectra of some components cannot be obtained, we develop “unknown components analysis (UCA),” whose calculation time is considerably less than those of the traditional algorithms. Compared to KCA, UCA requires more samples to be analyzed to ensure the high accuracy rate. These proposed algorithms provide new ways to detect and measure key biomedical substances.