In this study,serum Raman spectroscopy combined with pattern recognit ion technology was used to classify hepatitis B and C directly.Serum samples fr om 480hepatitis B patients and 346hepatitis C patients were measured for analy sis.The collected original spectra were preprocessed by baseline correction,an d then Primary Conponent Analysis (PCA) was used to reduce the spectral data dim ension to extract effective spectral data features.Finally,Support Vector Mach ine (SVM) and other algorithms were used to evaluate the difference between hepa titis B and C.This paper studies the support vector machine with two different optimization algorithms of Particle Swarm Optimization (PSO) and Grid Search (GS ) to classify hepatitis B patients and C patients.Additionally,Learning Vector Quantization (LVQ),Extreme Learning Machine (ELM),Linear discriminant Analysi s (LDA) and other pattern recognition algorithms were used to compare the classi fication of hepatitis B and C with the SVM model.In the current study,the PSO - SVM algorithm model achieved the best classification performance,and the accura cy of the model reached 96.74%. The results obtained in this work suggest that serum Raman spectroscopy combined with the PCA-PSO-SVM algorithm has great pot en tial for distinguishing between hepatitis B and C.This technique can be used to distinguish between hepatitis B and C.