A Universal and Accurate Method for Easily Identifying Components in Raman Spectroscopy Based on Deep Learning

Abstract

Raman spectroscopy has been widely used to provide the structural fingerprint for molecular identification. Due to interference from coexisting components, noise, baseline, and systematic differences between spectrometers, component identification with Raman spectra is challenging, especially for mixtures. In this study, a method entitled DeepRaman has been proposed to solve those problems by combining the comparison ability of a pseudo-Siamese neural network (pSNN) and the input-shape flexibility of spatial pyramid pooling (SPP). DeepRaman was trained, validated, and tested with 41,564 augmented Raman spectra from two databases (pharmaceutical material and S.T. Japan). It can achieve 96.29% accuracy, 98.40% true positive rate (TPR), and 94.36% true negative rate (TNR) on the test set. Another six data sets measured on different instruments were used to evaluate the performance of the proposed method from different aspects. DeepRaman can provide accurate identification results and significantly outperform the hit quality index (HQI) method and other deep learning models. In addition, it performs well in cases of different spectral complexity and low-content components. Once the model is established, it can be used directly on different data sets without retraining or transfer learning. Furthermore, it also obtains promising results for the analysis of surface-enhanced Raman spectroscopy (SERS) data sets and Raman imaging data sets. In summary, it is an accurate, universal, and ready-to-use method for component identification in various application scenarios.

Patch-Based Convolutional Encoder: A Deep Learning Algorithm for Spectral Classification Balancing the Local and Global Information

“…The backbone is stacked by four identical blocks whose internal structure is shown in Figure B. Eventually, by comparing the outcome, the global average pooling, instead of spatial pyramid pooling, is used by classifier to integrate all features and output probability distributions for each category via a fully connected layer (FC). The detailed processes of Patchify and the backbone are as follows.…”

A facile, low-cost bimetallic iron–nickel MOF nanozyme-propelled ratiometric fluorescent sensor for highly sensitive and selective uric acid detection and its smartphone application

“…56,57 Specifically, Fan et al proposed the concept of a DNA "contrary logic pair" (CLP) to try to reduce the time/cost of molecular computing based on functional nucleic acids. [58][59][60] After that, electrochemical and electro-chemiluminescent CLPs were further fabricated. 61,62 However, to the best of our knowledge, nanozyme-based CLP systems have barely been reported.…”

An interpretable and transferrable vision transformer model for rapid materials spectra classification

“…Recent advances in machine learning (ML), especially deep learning (DL), offer an exciting opportunity to reshape scientic research within the domains of chemical and materials science. [6][7][8] This is particularly evident in facilitating rapid analysis of intricate data, including but not limited to XRD, 9,10 IR/FTIR, 11,12 Raman, 13,14 and MS data. 15,16 For example, Oviedo and coworkers have demonstrated deployment of convolutional neural networks (CNNs) to effectively classify the dimensionalities and space groups of thin-lm metal halides from XRD spectra.…”