Application Researches On Articular Cartilage By Fourier Transform Infrared Imaging And Partial Least Squared Method

Articular cartilage is the functional connection of the joints, which plays an important role in people’s movement by reducing frictions and pressurments. Osteoarthritis is a kind of articular disease that was caused by the injury of joints, most patients that suffer pain will gradually lose their move ability. However, no special tool has been widely used in the diagnosing of early-stage osteoarthritis. In most cases, it could be identified only when the osteoarthritic degree becomes serious.Fourier transform infrared imaging(FTIRI) technique has been widely used in several different research fields, including material science, chemistry, pharmacy, biomedicine etc. It has also been used in the research of articular cartilage to get the spectral information and microscopic image of articular cartilage at same time.Chemometrics in spectral analysis can be used to quantitatively detect the concentration and distribution of macromolecules. It combining with FTIRI shows remarkable superiority in the content prediction of collagen and proteoglycan(PG) both in the healthy and osteoarthritic articular cartilage.This research aims to detect the collagen and PG content in bovine nasal cartilage, healthy and osteoarthritic articular cartilage, by using the combined method of FTIRI and partial least squared(PLS) regression. Moreover, a partial least squared discriminant analysis(PLS-DA) model was also built in this thesis, which was used to identify healthy and osteoarthritic articular cartilage. The main concerns and innovations of the paper are shown below:1. A standard spectral library was built for collagen and PG at first. PLS regression model was built based on the spectral library that related the infrared absorbances to macromolecular concentration. Additionally, several pre-processing methods were used to improve the stability of PLS model; leave-one-out cross validation was carried out to test the PLS model. Finally, bovine nasal cartilage samples were selected to verify the reliability and prediction ability of this PLS model.2. FTIR images of healthy and osteoarthritis articular cartilages were analyzed with PLS model, and then the concentration and distribution of collagen and PG were obtained. According to the PLS prediction, it was easily found that PG content was seriously lose in osteoarthritic articular cartilage when compared to the healthy samples, especially for the superficial zone and transitional zone.3. PLS-DA model was built on basis of PLS algorithm, which is an innovation. The discriminant accuracy was 100% and 90.24% for the calibration matrix and prediction matrix(18 spectrum), respectively. Based on our best knowledge and the previously research, this is the first time that PLS-DA method used for the discrimiant research of healthy and osteoarthritic articular cartilage. PLS-DA method might be a promising approach for the diagnosing of osteoarthritic disease.