In this study, a non-invasive approach based on surface-enhanced Raman spectroscopy (SERS) combined with machine learning is developed for the detection of lung cancer biomarker in blood serum at nanoscale level. Silver nanoparticles (Ag NPs) are synthesized using chemical method and characterized by UV-vis spectroscopy and Transmission electron microscopy (TEM) to conform their nanostructure. Moreover, the research aimed to differentiate the blood of healthy individuals from that of individuals with lung cancer. SERS spectra are analyzed and discussed using numerical results after comprehensive statistical and chemical analysis. The results showed significant changes in the spectral values and intensity of amino acids (phenylalanine), proteins (amide I and II), and lipids due to interaction between analyte and the metallic nanoparticles. Furthermore, a t-test demonstrated a broad statistical significance for the amide region (I) at an inverse frequency of 1655 cm⁻¹ with a p-value of 0.011255. Finally, after combining the results, a diagnostic accuracy of 95.74% is achieved, along with high sensitivity. By combining the SERS spectral fingerprint with machine learning and nanotechnology-based analysis, a robust and reliable result is obtained, enabling rapid and accurate identification of lung cancer. This approach extends the scope of the method to more medical applications.
Hamandi et al. (Fri,) studied this question.