ABSTRACT
Fourier transform infrared spectroscopic imaging (FTIRSI) technology can simultaneously obtain microstructure information and infrared spectral information of the samples. The method of FTIRSI combined with chemometric algorithms can be used for quantitative analysis of sample spectral information and tissue discrimination research. Based on this, FTIRSI and support vector machine classification (SVC) for the first time were used in this work to discriminate healthy and degenerated articular cartilage, with high accuracies of 100％ and 95. 4％ , respectively, and sum accuracy of 97. 7％ . The support vector regression (SVR) model was used to quantitatively study the contents and distribution of two biomacromolecules, collagen and proteoglycan, in articular cartilage. The proteoglycan loss occurred in the degenerated articular cartilage, especially in the superficial area. This study indicates that the combination of FTIRSI and support vector machine (SVM) is expected to become a new diagnostic tool for osteoarthritis, which is of great significance for the early diagnosis and research of osteoarthritis.