In this study, we investigate the biochemical characteristics of oxidative stress in age-related macular degeneration (AMD) and diabetic retinopathy (DR) by analyzing aqueous humors. Nondiabetic cataract aqueous humor was used as the control. The level of oxidative damage was evaluated based on changes in Raman spectral intensity. Seven prominent peaks were detected at 1002, 1043, 1062, 1352, 1419, 1454, and 1656 cm . We proposed four multimodal biomarkers to distinguish these peaks based on the ratios of Raman intensities in two wavelengths, including CHO (C-O stretching or C-O-H bending modes), AG (adenine and guanine), PRO-AG (protein and AG), and PHEα (phenylalanine symmetric ring breath and amide I α-helix) markers. The presence of oxidative damage was detected by CHO and AG markers associated with C-O stretching, C-O-H bending modes in carbohydrates (1043 cm ), and the nucleic acids adenine and guanine (1352 cm ), respectively. DR-related oxidative damage was identified by PRO-AG and PHEα markers associated with adenine, guanine, and protein components (1419 and 1454 cm ) and amide I α-helix protein structure (1656 cm ), respectively. AMD-related oxidative damage was identified by four biomarkers. Four multimodal biomarkers with simple linear threshold values achieved high sensitivity of 100% and high specificity of 100% for classifying oxidative stress-induced AMD and DR diseases. Therefore, Raman-based label-free optical detection is effective for detecting the presence of age-related or diabetic oxidative damage in aqueous humor.