Serum-based surface-enhanced Raman spectroscopy combined with PCA-RCKNCN for rapid and accurate identification of lung cancer.

Early and precise diagnosis of lung cancer is critical for a better prognosis. However, it is still a challenge to develop an effective strategy for early precisely diagnose and effective treatments. Here, we designed a label-free and highly accurate classification serum analytical platform for identifying mice with lung cancer. Specifically, the microarray chip integrated with Au nanostars (AuNSs) array was employed to measure the surface-enhanced Raman scattering (SERS) spectra of serum of tumor-bearing mice at different stages, and then a recognition model of SERS spectra was constructed using the principal component analysis (PCA)-representation coefficient-based k-nearest centroid neighbor (RCKNCN) algorithm. The microarray chip can realize rapid, sensitive, and high-throughput detection of SERS spectra of serum. RCKNCN based on the PCA-generated features successfully differentiated the SERS spectra of serum of tumor-bearing mice at different stages with a classification accuracy of 100%. The most prominent spectral features for distinguishing different stages were captured in PCs loading plots. This work not only provides a practical SERS chip for the application of SERS technology in cancer screening, but also provides a new idea for analyzing the feature of serum at the spectral level. [Display omitted] • A rapid, sensitive, label-free, high-throughput SERS microarray chip was developed. • Combining SERS with PCA-RCKNCN successfully differentiated the SERS spectra. • The most prominent spectral features of SERS spectra in PCs loading were captured. • PCA-RCKNCN was superior to traditional multivariate algorithm in accuracy. [ABSTRACT FROM AUTHOR]