Abstract

Biomarkers for acute myocarditis (AM) are commonly used in clinical diagnosis, but their ultralow concentrations in blood are challenging to be detected in the early stage. This study presents a novel bowl-shaped surface-enhanced Raman scattering (SERS) substrate with high-performance, prepared via one-step hydrofluoric acid etching of SiO 2 microspheres, 515 nm femtosecond-laser reduction of gold ions, and nanosecond-laser reshaping of gold nanoparticles (AuNPs). In particular, 500 nm SiO 2 microspheres were etched with a hydrofluoric acid solution to form a uniform microbowl array, and AuNPs were synthesized in situ on the SiO 2 microbowls via irradiation with a femtosecond laser. Subsequently, the AuNPs were reshaped into nanorods using a nanosecond laser. The resulting bowl-shaped AuNP@SiO 2 substrate showed a SERS enhancement factor of 10 6 and a limit of detection of 3.84 × 10 −8 M, as evaluated on Rhodamine 6G. Notably, the limits of detection of the AM biomarkers were lower than 6 pg/mL, enabling the accurate detection of their trace quantities. For AM prediction, a fuzzy logic model was trained, enabling the accurate and quantitative analysis of biomarkers, with a mean relative error of less than 25.7%. Thus, the proposed SERS platform is a reliable tool for the early noninvasive diagnosis of AM.