时域的磁共振波谱(magnetic resonance spectroscopy,MRS)信号经过傅立叶变换得到频域波谱,通常我们仅对某段频率范围内的波谱感兴趣。本文采用基于奇异值分解(singular value decomposition,SVD)和频率选择的方法,仅对感兴趣的频率范围内的波谱进行量化,降低了其它频率的波谱对感兴趣区域的影响,并大大减少了运算量。通过与时域标准的SVD方法比较实验,表明该算法准确、省时。 Time-domain magnetic resonance spectroscopy (magnetic resonance spectroscopy, MRS) signal through Fourier transform to get the frequency domain spectrum, usually we only interested in a certain frequency range of the spectrum. In this paper, based on singular value decomposition (SVD) and frequency selection method, only the spectrum in the frequency range of interest is quantified to reduce the influence of the spectrum of other frequencies on the region of interest and greatly reduce the computation the amount. By comparing with the time domain standard SVD method, it shows that the algorithm is accurate and time-saving.