A spatial-spectral feature fusion attention residual network for automated bacterial classification via atomic force microscopy

Key Points

• Automated classification achieves high accuracy with reduced manual intervention, enhancing diagnostics.
• Key metric shows over 90% accuracy in identifying bacterial species across multiple tests.
• Analysis incorporates a spatial-spectral feature fusion attention residual network for optimal results.
• These findings suggest promising applications for microbiology, though validation in wider datasets is needed.