Protocol for cerebrospinal fluid analysis using enrichment-enhanced surface-enhanced Raman spectroscopy and transformer-enabled spectral classification

The rapid detection and precise classification of cerebrospinal fluid in acute leukemia patients constitute a crucial clinical imperative. Here, we present a protocol for cerebrospinal fluid analysis deep learning with enrichment-enhanced surface-enhanced Raman spectroscopy (DL-SERS) detection and transformer-enabled spectral classification. We describe steps for preparing SERS-active materials and samples and acquiring SERS spectral data. We then detail procedures for deep learning-based classification. This protocol enables highly sensitive and efficient identification of acute leukemia patients. For complete details on the use and execution of this protocol, please refer to Zhang et al. 1