エトミデート汚染指紋の化学的視覚化と照合のための超音波支援SERSイメージング

The inadvertent deposition of latent fingerprints stamps indubitable forensic evidence, serving as a solid link between individuals and objective scenes. In this study, a novel approach was developed to simultaneously collect the geometrical and chemical features from latent fingerprints, which establishes a broadened perspective over the inspected biological traces. Ultrasonic spraying was employed to swiftly acquire a uniform distribution of noble metal nanoparticles (NPs) for surface-enhanced Raman spectroscopy (SERS) imaging. The NP-developed fingerprint brings an improved contrast of microscopic visualization as well as additional inelastic spectral information within the SERS image. K-NearestNeighbor (KNN) feature point matching and local similarity matching were utilized to recognize the SERS fingerprint images reconstructed from characteristic SERS bands. A matching score involving minutia descriptors was introduced to parametrize the degree of resemblance in latent fingerprint identification, yielding high matching accuracy even with partially damaged fingerprints. The trade-off between the accuracy in chemical identification and the completeness in fingerprint reconstruction was discussed. The effective SERS visualization and reconstruction of fingerprints was accomplished from the sample containing etomidate with the concentration as low as 1 μM. Additionally, the SERS imaging on a curved keycap surface was realized, which meets the critical demands of the investigation over particular chemicals.