ABSTRACT
Surface-enhanced Raman spectroscopy (SERS), a rapid, low-cost, non-invasive, ultrasensitive, and label-free technique, has been widely used in-situ and ex-situ biomedical diagnostics questions. However, analyzing and interpreting the untargeted spectral data remains challenging due to the difficulty of designing an optimal data pre-processing and modelling procedure. In this paper, we propose a Multi-branch Attention Raman Network (MBA-RamanNet) with a multi-branch attention module, including the convolutional block attention module (CBAM) branch, deep convolution module (DCM) branch, and branch weights, to extract more global and local information of characteristic Raman peaks which are more distinctive for classification tasks. CBAM, including channel and spatial aspects, is adopted to enhance the distinctive global information on Raman peaks. DCM is used to supplement local information of Raman peaks. Autonomously trained branch weights are applied to fuse the features of each branch, thereby optimizing the global and local information of the characteristic Raman peaks for identifying diseases. Extensive experiments are performed for two different neurological disorders classification tasks via untargeted serum SERS data. The results demonstrate that MBA-RamanNet outperforms commonly used CNN methods with an accuracy of 88.24% for the classification of healthy controls, mild cognitive impairment, Alzheimer's disease, and Non-Alzheimer's dementia; an accuracy of 90% for the classification of healthy controls, elderly depression, and elderly anxiety.
ABSTRACT
Identification of age-related neuropsychiatric disorders, i.e., late-life depression (LDD) and mild cognitive impairment (MCI) is of imperative clinical value considering the large probability of misdiagnosis and current lack of sensitive, non-invasive and low-cost diagnostic approaches. Here, the serum surface-enhanced Raman spectroscopy (SERS) technique is proposed to identify healthy controls, LDD and MCI patients. Based on SERS peaks analysis, abnormal levels of ascorbic acid, saccharide, cell-free DNA and amino acids in serum are found to be potential biomarkers for identifying LDD and MCI. These biomarkers might be related to oxidative stress, nutritional status, lipid peroxidation and metabolic abnormalities. Moreover, partial least square analysis-linear discriminant analysis (PLS-LDA) is applied to those collected SERS spectra. Finally, the overall identification accuracy is 83.2%, and accuracies are 91.6% and 85.7% for differentiating healthy versus neuropsychiatric disorders and LDD versus MCI, respectively. Thus, the serum SERS combined with multivariate statistical analysis has proved its successful potential for rapid, sensitive and non-invasive identification of healthy, LDD and MCI, which may open new avenues for early diagnosis and timely intervention for age-related neuropsychiatric disorders.
