Label-Free Detection and Classification of Glaucoma Based on Drop-Coating Deposition Raman Spectroscopy

Primary open-angle glaucoma (POAG) and primary angle-closure (PACG) are prevailing eye diseases that can lead to blindness. In order provide a non-invasive diagnostic method for glaucoma, we investigated the feasibility of using drop-coating deposition Raman spectroscopy (DCDRS) discriminate patients from healthy individuals based on tear samples. Tears 27, 19 27 POAG patients, PACG normal individuals, respectively, were collected measurement. For high-dimension data analysis, principal component analysis–linear discriminant analysis (PCA-LDA) was used features spectra, followed by support vector machine (SVM) classify samples into three categories, which is called PCA-LDA-based SVM. The differences in characteristic peaks spectra between people related different contents various proteins lipids. SVM, total accuracy reached 93.2%. With evaluation 30% test dataset validation, classification model 90.9%. results this work reveal tears be detection discrimination combining process with supporting DCDRS being potential diagnosis future.