Anterior cruciate ligament (ACL) plays an essential role in knee joint stability and kinematics. The microstructural irregularities such as cellular changes and disorganization of the extracellular matrix (ECM) alter the mechanical properties of the ligament, leading to a significant knee functional instability and progression of osteoarthritis (OA). So far, the identification of the local abnormality in ACL has routinely relied on invasive analytical techniques such as histology or biochemical assays. The non-invasive diagnosis using magnetic resonance imaging (MRI) is still limited to identifying the presence/absence of partial/complete ruptures and mucoid degeneration. In this study, laser micro-Raman spectroscopy with near-infrared excitation (785 nm) was applied to human ACL in order to establish optical algorithms for non-destructively diagnosing a degeneration state at molecular level. Raman spectra were obtained from 44 ex-vivo ACL specimens, and these were subsequently classified as an early (subclinical) and advanced (clinical) level of tissue degradation based on the histopathological scoring system. The significant differences in Raman peak intensities were found between the different degeneration groups, which were assigned to the vibrational modes of nucleic acids in cells, collagens, and phospholipids. Linear discriminant analysis (LDA) was performed to identify cut-off values for the distributions of Raman intensity and intensity ratios, which enable to best discriminate between the early and advanced degenerated tissues. Raman intensity algorithms derived from I1101/I1749, [I1002/I1516vs. I1101/I1749], and [I1002/I1749vs. I1101/I1749], yielded a maximum diagnostic sensitivity of 100%, specificity of 80%, and accuracy of 91% for discriminating the degeneration severity. STATEMENT OF SIGNIFICANCE: In this study, laser micro-Raman spectroscopy was applied to human anterior cruciate ligament (ACL) to establish optical algorithms for non-destructively diagnosing the tissue degeneration at molecular level. To our knowledge, this is the first report on Raman diagnosis for human ACL. Linear discriminant analysis (LDA) was performed to identify cut-off values for Raman intensity and intensity ratios, which enable to best discriminate between an early (subclinical) and advanced (clinical) level of ACL degeneration. The intensity ratios of I1101/I1749, [I1002/I1516vs. I1101/I1749], and [I1002/I1749vs. I1101/I1749] yielded a maximum diagnostic sensitivity of 100%, specificity of 80%, and accuracy of 91% for discriminating the ACL degeneration. The present findings might contribute to expanding clinical diagnostic possibilities for non-invasively identifying tissue degeneration.