Translator Disclaimer
9 August 2018 A multiclass classification method for building MALDI-TOF mass spectrometry database of microbes based on SVM
Author Affiliations +
Proceedings Volume 10806, Tenth International Conference on Digital Image Processing (ICDIP 2018); 108065L (2018) https://doi.org/10.1117/12.2503120
Event: Tenth International Conference on Digital Image Processing (ICDIP 2018), 2018, Shanghai, China
Abstract
This study proposes a multiclass classification method for identifying microbes based on MALDI-TOF-MS that is developed specifically for the characterization of microbial protein fingerprint compared to the other technologies. The raw data of mass charge ratio are distinguished from each other for the two dimensions of the ratio and its matched signal intensity. After deleting the noise signal using moving average method and screening out the peaks signal by filtering selected local extremum, we trained support vector machine in one-versus-one and one-versus-rest mode to build a scoring mechanism. By taking Euclidean Distance inside, a formula was furthermore established to judge the unknown samples, which can serve as an effective method to assist the database construction of microbes with reliable results.
© (2018) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Jianqi Wu, Yang Yu, and Weidong Pan "A multiclass classification method for building MALDI-TOF mass spectrometry database of microbes based on SVM", Proc. SPIE 10806, Tenth International Conference on Digital Image Processing (ICDIP 2018), 108065L (9 August 2018); https://doi.org/10.1117/12.2503120
PROCEEDINGS
6 PAGES


SHARE
Advertisement
Advertisement
RELATED CONTENT


Back to Top