Rapid and accurate identification of pathogenic bacteria at the single‐cell level using laser tweezers Raman spectroscopy and deep learning

Abstract

We report a new method for the rapid identification of pathogenic bacterial species at the single‐cell level that combines laser tweezers Raman spectroscopy (LTRS) with deep learning (DL). LTRS can accurately measure single‐cell Raman spectra (scRS) without destroying and labeling cells. Based on the scRS data, DL rapidly and accurately identifies pathogenic bacteria. We measured scRS of 15 species bacteria using homemade LTRS. For each species, approximately, 160 cells from three different patients were measured, one patient's data were used as test set, and the rest after being augmented was used as training set. A residual network (ResNet) model, trained on the augmented training set, achieved an accuracy of 94.53% on the test set. Moreover, we applied gradient‐weighted class activation mapping to visualize the proposed model. Finally, we demonstrated the advantages of ResNet over traditional machine‐learning algorithms.

RamanNet: a lightweight convolutional neural network for bacterial identification based on Raman spectra

“…We also tested RamanNet on another pathogenic bacterial dataset, the PKU-bacterial dataset, to demonstrate its robustness. The PKU-bacterial dataset was established by ourselves in 2022, 15 which contains single-cell Raman spectra (scRS) of 15 pathogenic bacteria species. For each species, approximately 160 cells were isolated from three different patients; one patient's data were used as the test set, and the data from the other two patients were first augmented and then used as the training set.…”

“…10,14 In recent years, the use of neural networks to automatically feature Raman spectra has significantly improved the identification accuracy of pathogenic bacteria compared to traditional machine learning algorithms. 8,9,15…”

“…10,14 In recent years, the use of neural networks to automatically feature Raman spectra has signicantly improved the identication accuracy of pathogenic bacteria compared to traditional machine learning algorithms. 8,9,15 To improve the classication accuracy of convolutional neural network (CNN) for bacterial Raman spectra, many attempts have focused on using dozens of lters, 14,[16][17][18] increasing the depth of the network, 14,16 and adopting more advanced classication algorithm models, which were mainly used in the optimization methods of the image classication eld. These methods oen result in longer training times and more complex hyperparameter designs for neural networks.…”

The scientific career and contributions of Prof. Wang Yiqiu

“…For instance, [133] explored one-dimension convolutional neural networks (1CNNs) and applied them to identify marine microbes with single-cell Raman spectroscopy. [134] furtherly developed the CNN model using the residual network (ResNet) model and succeeded in rapidly and accurately identifying 15-species pathogenic bacteria with the accuracy of 94.53% at the singlecell level.…”

Visualization of a Machine Learning Framework toward Highly Sensitive Qualitative Analysis by SERS

“…However, due to the small computation of these algorithms, their performance is less than satisfactory in samples where strong molecular interaction was involved, which was partly solved with the introduction of machine learning. For example, early diagnosis, , bacterial and related drug resistance, and quality control of edible oils have been realized by deep neural networks.…”