Bacteria Raman spectrum identification and classification method based on SIMCA-SVDD

Abstract

The invention discloses a bacterium Raman spectrum recognition and classification method based on SIMCA-SVDD, and belongs to the field of bacterium Raman spectrum analysis. The method comprises the following steps: firstly, carrying out data preprocessing on collected bacteria Raman data, then inputting the preprocessed data into an SIMCA-SVDD algorithm for modeling, and then carrying out prediction classification through bacteria Raman test set data. According to the SIMCA-SVDD algorithm, SVDD maps original data to a high-dimensional space, and an original nonlinear inseparable problem is converted into linear separability in the high-dimensional space, so that a smaller division area range can be obtained and contains more target bacterium Raman spectrum sample information, and a betterclassification result is obtained; the verification is carried out through the acquired Raman spectrum data of the four types of different bacteria; it is shown that the classification accuracy of escherichia coli, shigella flexneri, clostridium difficile, SIMCA and SIMCASVDD is 100%, but the classification accuracy of clostridium botulinum, SIMCA and SIMCASVDD is 86.7% and 93.3% respectively. Asa result, it can be seen that SIMCASVDD has certain superiority in identification of clostridium botulinum.

Description

technical field [0001] The invention relates to a rapid identification and classification method for bacterial Raman spectrum, and relates to the field of bacterial Raman spectrum analysis. Background technique [0002] Clostridium botulinum (Clostridium botulinum) is an anaerobic bacterium that can easily survive in a closed anaerobic environment. As an important pathogen causing bacterial poisoning worldwide, Clostridium botulinum has always been a key monitoring object in the field of disease control. Clostridium botulinum is widely found in all kinds of vacuum-packed foods including canned and frozen foods. Likewise, if Clostridium botulinum is present in a water supply, it can cause more extensive damage. [0003] In the past few decades, the field of bacterial identification and detection has produced many new ideas to solve problems, including fluorescence in situ hybridization, mass spectrometry, polymerase chain reaction, etc. These methods are time-consuming and...

AI Technical Summary

Problems solved by technology

The traditional method also has its defects: for example, the test results cannot realize the final identification of the strains, and the detection accuracy is not high, etc.

The original SIMCA method uses Euclidean distance to classify samples. For the nonlinear problems encountered, the circle cannot describe the boundary of samples very accurately.

Because the substances contained in bacteria are similar, the spectra of bacterial spectra are relatively similar, and the classification of sample data itself is difficult.

Images