Rapid identification of alpha-fetoprotein in serum by a microfluidic SERS chip integrated with Ag/Au Nanocomposites

• A six-channel microfluidic chip integrated with automatic injection and the Ag/Au nanocomposites SERS enhanced substrate was designed. • SERS analytical methodology and statistical methods were combined to establish an efficient and rapid alpha-fetoprotein detection method in serum. • For hepatocellular carcinoma patients’ serum samples and normal serum samples, the classification accuracy by the proposed PCA-LDA model was 96.25% and the blind sample test accuracy was 95%. The rapid detection of alpha-fetoprotein (AFP) in serum is of great significance in the early diagnosis of hepatocellular carcinoma (HCC). A novel six-channel microfluidic SERS chip integrated with Ag/Au nanocomposites (NCs) for rapid SERS identification of AFP in serum was proposed in this work. Automatic injection was realized by negative pressure formed by specially designed PDMS module on the SERS chip. The serum samples originating from 60 HCC patients and 60 healthy people were measured by SERS on the designed microchip under the optimized conditions. The SERS spectra of serum samples from normal people and patients were collected and analyzed by Principal Component Analysis (PCA) which separated the characteristic Raman peaks of AFP of the two groups into two distinct clusters. Linear Discriminate Analysis (LDA) based on the PCA generated features differentiated the patients’ sera SERS spectra from the normal sera SERS spectra with a classification accuracy of 96.25% and a blind sample test accuracy of 95%. It is shown that the proposed microfludic SERS chip for AFP SERS test in serum can rapidly and efficiently identify HCC patients. It is of great research value and practical prospect in the fields of disease diagnosis and screening.