Probabilistic neural network-based classifier of ToF-SIMS single-pixel spectra

Nunzio Tuccitto; Alessandra Bombace; Alberto Torrisi; Antonino Licciardello; Grazia Lo Sciuto; Giacomo Capizzi; Marcin Woźniak

doi:10.1016/j.chemolab.2019.07.002

Probabilistic neural network-based classifier of ToF-SIMS single-pixel spectra

Nunzio Tuccitto
, Alessandra Bombace
, Alberto Torrisi
, Antonino Licciardello
, Grazia Lo Sciuto
, Giacomo Capizzi
, Marcin Woźniak

Department of Artificial Intelligence Modelling (RMS/RMS2)

University of Catania

Research output: Contribution to journal › Article › peer-review

16 Citations (Scopus)

Abstract

A new automated procedure able to extract latent chemical information from ToF-SIMS big data sets is presented. A classifier based on a probabilistic neural network able to provide a correct classification of spectra acquired from four different polymers is designed and trained. The procedure is fast and low-demanding in terms of CPU performances, and it is also able to evaluate the similarity/dissimilarity in the Fourier transform domain of very low intensity single pixel spectra without any effort of the analyst.

Original language	English
Pages (from-to)	138-142
Number of pages	5
Journal	Chemometrics and Intelligent Laboratory Systems
Volume	191
DOIs	https://doi.org/10.1016/j.chemolab.2019.07.002
Publication status	Published - 15 Aug 2019

Keywords

Feature extraction
Neural networks
Secondary ion mass spectrometry
ToF-SIMS single-pixel spectra classification
ToF-SIMS spectra

ASJC Scopus subject areas

Analytical Chemistry
Software
Process Chemistry and Technology
Spectroscopy
Computer Science Applications

Access to Document

10.1016/j.chemolab.2019.07.002

Cite this

@article{4388f90dff5844ec84d0b7e68aef727b,

title = "Probabilistic neural network-based classifier of ToF-SIMS single-pixel spectra",

abstract = "A new automated procedure able to extract latent chemical information from ToF-SIMS big data sets is presented. A classifier based on a probabilistic neural network able to provide a correct classification of spectra acquired from four different polymers is designed and trained. The procedure is fast and low-demanding in terms of CPU performances, and it is also able to evaluate the similarity/dissimilarity in the Fourier transform domain of very low intensity single pixel spectra without any effort of the analyst.",

keywords = "Feature extraction, Neural networks, Secondary ion mass spectrometry, ToF-SIMS single-pixel spectra classification, ToF-SIMS spectra",

author = "Nunzio Tuccitto and Alessandra Bombace and Alberto Torrisi and Antonino Licciardello and \{Lo Sciuto\}, Grazia and Giacomo Capizzi and Marcin Wo{\'z}niak",

note = "Publisher Copyright: {\textcopyright} 2019 Elsevier B.V.",

year = "2019",

month = aug,

day = "15",

doi = "10.1016/j.chemolab.2019.07.002",

language = "English",

volume = "191",

pages = "138--142",

journal = "Chemometrics and Intelligent Laboratory Systems",

issn = "0169-7439",

publisher = "Elsevier B.V.",

}

TY - JOUR

T1 - Probabilistic neural network-based classifier of ToF-SIMS single-pixel spectra

AU - Tuccitto, Nunzio

AU - Bombace, Alessandra

AU - Torrisi, Alberto

AU - Licciardello, Antonino

AU - Lo Sciuto, Grazia

AU - Capizzi, Giacomo

AU - Woźniak, Marcin

PY - 2019/8/15

Y1 - 2019/8/15

N2 - A new automated procedure able to extract latent chemical information from ToF-SIMS big data sets is presented. A classifier based on a probabilistic neural network able to provide a correct classification of spectra acquired from four different polymers is designed and trained. The procedure is fast and low-demanding in terms of CPU performances, and it is also able to evaluate the similarity/dissimilarity in the Fourier transform domain of very low intensity single pixel spectra without any effort of the analyst.

AB - A new automated procedure able to extract latent chemical information from ToF-SIMS big data sets is presented. A classifier based on a probabilistic neural network able to provide a correct classification of spectra acquired from four different polymers is designed and trained. The procedure is fast and low-demanding in terms of CPU performances, and it is also able to evaluate the similarity/dissimilarity in the Fourier transform domain of very low intensity single pixel spectra without any effort of the analyst.

KW - Feature extraction

KW - Neural networks

KW - Secondary ion mass spectrometry

KW - ToF-SIMS single-pixel spectra classification

KW - ToF-SIMS spectra

UR - https://www.scopus.com/pages/publications/85068973972

U2 - 10.1016/j.chemolab.2019.07.002

DO - 10.1016/j.chemolab.2019.07.002

M3 - Article

AN - SCOPUS:85068973972

SN - 0169-7439

VL - 191

SP - 138

EP - 142

JO - Chemometrics and Intelligent Laboratory Systems

JF - Chemometrics and Intelligent Laboratory Systems

ER -

Probabilistic neural network-based classifier of ToF-SIMS single-pixel spectra

Abstract

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this