Practical SERS substrates by spray coating of silver solutions for deep learning-assisted sensitive antigen identification

Basit Öğe Kaydı
dc.authorid0000-0002-2261-1545
dc.authorid0000-0001-5409-3925
dc.contributor.authorSahin, Furkan
dc.contributor.authorSahin, Gamze Demirel
dc.contributor.authorCamdal, Ali
dc.contributor.authorAkmayan, Ilkgul
dc.contributor.authorOzbek, Tulin
dc.contributor.authorAcar, Serap
dc.contributor.authorOnses, Mustafa Serdar
dc.date.accessioned2026-01-31T15:08:16Z
dc.date.available2026-01-31T15:08:16Z
dc.date.issued2025
dc.departmentİstanbul Beykent Üniversitesi
dc.description.abstractSurface-enhanced Raman spectroscopy (SERS) has long been recognized for its rapid and sensitive detection capabilities; however, challenges persist in practical fabrication of the substrates and interpreting complex data. Herein, we propose a deep learning (DL) assisted SERS approach to enable rapid and sensitive detection of analytes on practical yet highly effective substrates prepared by direct spray-coating of a nanoparticle-free true solution of a reactive Ag ink and on-site thermal annealing mediated generation of nanostructures. This design ensured homogeneous distribution of Ag nanostructures throughout the entire substrate, significantly increasing the number of hotspots and enhancing the Raman signals, thereby achieving an impressive analytical enhancement factor of similar to 10(10) in a reproducible and consistent manner. The diagnostic utility of this platform was demonstrated by detecting the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) spike (S) protein in both buffer and saliva, with detection limits of 74.3 pg/mL and 7.43 ng/mL, respectively. The DL-assisted SERS not only accurately identified the presence or absence of viral antigen, but also automatically quantified the viral load. This automatic identification achieved an outstanding accuracy of similar to 99.9 %, highlighting the exceptional performance of the proposed platform. This simple, cost-effective, scalable, and ultra-sensitive DL-assisted SERS platform offers significant opportunities for early and precise detection in a range of analytical scenarios.
dc.description.sponsorshipResearch Fund of the Erciyes University [FOA-2023-12834]
dc.description.sponsorshipThis work was supported by the Research Fund of the Erciyes University (Project Number FOA-2023-12834) .
dc.identifier.doi10.1016/j.colsurfa.2024.135828
dc.identifier.issn0927-7757
dc.identifier.issn1873-4359
dc.identifier.urihttps://doi.org./10.1016/j.colsurfa.2024.135828
dc.identifier.urihttps://hdl.handle.net/20.500.12662/10642
dc.identifier.volume707
dc.identifier.wosWOS:001439838400001
dc.identifier.wosqualityQ2
dc.indekslendigikaynakWeb of Science
dc.language.isoen
dc.publisherElsevier
dc.relation.ispartofColloids And Surfaces A-Physicochemical And Engineering Aspects
dc.relation.publicationcategoryMakale - Uluslararası Hakemli Dergi - Kurum Öğretim Elemanı
dc.rightsinfo:eu-repo/semantics/closedAccess
dc.snmzKA_WoS_20260128
dc.subjectSERS
dc.subjectDeep-learning
dc.subjectAg ink
dc.subjectSpray coating
dc.subjectViral load
dc.subjectVirus antigen
dc.titlePractical SERS substrates by spray coating of silver solutions for deep learning-assisted sensitive antigen identification
dc.typeArticle

Dosyalar

Koleksiyon

WoS İndeksli Yayınlar Koleksiyonu