The human carcinogenicity evaluation of chemicals has a great impact on public health. In vitro methods, such as the cell transformation assay (CTA), allow for a fast and reliable assessment of the carcinogenic potential of a chemical compound in comparison with the standard two-year bioassay. The scoring and classification of foci in selected cell lines is performed, after staining, by light microscopy. Foci can be separated into three classes: type I, which are scored as non-transformed, and types II and III that are considered to include fully transformed foci. However, in a number of cases, even an expert is uncertain about the attribution of a focus to a given class, due to its mixed or intermediate nature. Here, we suggest a simple approach to classifying mixed or intermediate foci by exploiting the quantitative information available from images, which is captured by statistical descriptors. A quantitative index is proposed, to describe the degree of dissimilarity of mixed and intermediate images to the three well-distinguished classes.

Image classifiers for the cell transformation assay: a progress report / C. Urani, G.F. Crosta, C. Procaccianti, P. Melchioretto, F.M. Stefanini (PROCEEDINGS OF SPIE, THE INTERNATIONAL SOCIETY FOR OPTICAL ENGINEERING). - In: Imaging, Manipulation, and Analysis of Biomolecules, Cells, and Tissues VIII / [a cura di] D.L. Farkas, D.V. Nicolau, R.C. Leif. - [s.l] : SPIE, 2010. - ISBN 9780819479648. - pp. 1-11 (( convegno Conference on Imaging, Manipulation, and Analysis of Biomolecules, Cells, and Tissues VII tenutosi a San Francisco nel 2010.

Image classifiers for the cell transformation assay: a progress report

F.M. Stefanini
2010

Abstract

The human carcinogenicity evaluation of chemicals has a great impact on public health. In vitro methods, such as the cell transformation assay (CTA), allow for a fast and reliable assessment of the carcinogenic potential of a chemical compound in comparison with the standard two-year bioassay. The scoring and classification of foci in selected cell lines is performed, after staining, by light microscopy. Foci can be separated into three classes: type I, which are scored as non-transformed, and types II and III that are considered to include fully transformed foci. However, in a number of cases, even an expert is uncertain about the attribution of a focus to a given class, due to its mixed or intermediate nature. Here, we suggest a simple approach to classifying mixed or intermediate foci by exploiting the quantitative information available from images, which is captured by statistical descriptors. A quantitative index is proposed, to describe the degree of dissimilarity of mixed and intermediate images to the three well-distinguished classes.
image classification; neoplastic transformation; quantitative morphology; spectrum enhancement; multivariate statistics; image understanding; unsupervised training; histogram moments
Settore SECS-S/01 - Statistica
Book Part (author)
File in questo prodotto:
Non ci sono file associati a questo prodotto.
Pubblicazioni consigliate

Caricamento pubblicazioni consigliate

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: http://hdl.handle.net/2434/849482
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 0
  • ???jsp.display-item.citation.isi??? 0
social impact