Non-invasive detection of systemic lupus erythematosus using SERS serum detection technology and deep learning algorithms

• Published in: Spectrochimica acta. Part A, Molecular and biomolecular spectroscopy Vol. 320; p. 124592
• Main Authors: Wang, Xuehua, Hou, Junwei, Chen, Chen, Jia, Zhenhong, Zuo, Enguang, Chang, Chenjie, Huang, Yuhao, Chen, Cheng, Lv, Xiaoyi
• Format: Journal Article
• Language: English
• Published: England Elsevier B.V 05.11.2024
• Subjects: Adult; Algorithms; Au@Ag-PSi substrate; Classification diagnosis; Deep Learning; Deep learning algorithm; Female; Gold - chemistry; Humans; Lupus Erythematosus, Systemic - blood; Lupus Erythematosus, Systemic - diagnosis; Metal Nanoparticles - chemistry; Rhodamines - chemistry; Silicon - chemistry; Silver - chemistry; Spectrum Analysis, Raman - methods; Surface Enhanced Raman Scattering (SERS); Systemic lupus erythematosus (SLE); Systemic lupus erythematosus (SLE); Surface Enhanced Raman Scattering (SERS); Classification diagnosis; Au@Ag-PSi substrate; Deep learning algorithm
• ISSN: 1386-1425; 1873-3557
• DOI: 10.1016/j.saa.2024.124592

[Display omitted] •A novel sea urchin-like chip with Au NPs as the shell, Ag NPs as the core and PSi as the substrate was synthesized by electrochemical etching and in situ reduction.•Three models, CNN, AlexNet, and RF, were selected to model and analyze the serum spectra of SLE patients and healthy controls, and the model classification accuracy reached 92 %.•A fast, accurate, and non-invasive screening method for SLE was established by combining SERS technology and deep learning algorithms. Systemic lupus erythematosus (SLE) is an autoimmune disease with multiple symptoms, and its rapid screening is the research focus of surface-enhanced Raman scattering (SERS) technology. In this study, gold@silver-porous silicon (Au@Ag-PSi) composite substrates were synthesized by electrochemical etching and in-situ reduction methods, which showed excellent sensitivity and accuracy in the detection of rhodamine 6G (R6G) and serum from SLE patients. SERS technology was combined with deep learning algorithms to model serum features using selected CNN, AlexNet, and RF models. 92 % accuracy was achieved in classifying SLE patients by CNN models, and the reliability of these models in accurately identifying sera was verified by ROC curve analysis. This study highlights the great potential of Au@Ag-PSi substrate in SERS detection and introduces a novel deep learning approach for SERS for accurate screening of SLE. The proposed method and composite substrate provide significant value for rapid, accurate, and noninvasive SLE screening and provide insights into SERS-based diagnostic techniques.