In　clinical　practice,　infections　caused　by　Aspergillus　species　are　common　and　associated　with　high　mortality　rates.　However,　isolating　Aspergillus　from　clinical　samples　and　cultures　is　challenging,　and　serological　tests　cannot　differentiate　between　strains.　Furthermore,　the　complex　cell　wall　structure　in　Aspergillus　species　hinders　DNA　extraction　and　subsequent　nucleic　acid　detection.　Aspergillus　fumigatus,　Aspergillus　flavus,　and　Aspergillus　niger　are　common　and　frequently　encountered　pathogenic　species.　This　study　aimed　to　establish　an　efficient　and　rapid　method　for　detecting　Aspergillus　species　in　clinical　samples.　Fe3O4@PEI.NH2　nanoparticles　with　large　particle　sizes　were　employed　to　capture　Aspergillus　from　the　samples.　Flexible　silver　nanowires　(AgNWs)　were　prepared　and　used　as　the　surface-enhanced　Raman　scattering　(SERS)　substrate　for　Aspergillus　detection.　Through　SERS　spectroscopy,　the　identification　of　captured　Aspergillus　species　was　confirmed.　Furthermore,　the　SERS　spectrum　underwent　principal　component　analysis　employing　orthogonal　partial　least-squares　discrimination　analysis　(OPLS-DA)　as　a　multivariate　analysis　technique　to　distinguish　between　the　various　Aspergillus　species　effectively.　After　10-fold　cross-validation,　the　trained　model　achieved　a　testing　accuracy　of　99.15%,　indicating　excellent　classification　performance.　By　utilizing　pre-prepared　Fe3O4@PEI.NH2　and　AgNWs,　the　detection　process　was　completed　within　60　min　without　disrupting　the　cell　wall　of　Aspergillus.　©　2023　The　Authors