Abstract

RODRIGUEZ GUILLEN, Reinier; GARCIA GARCIA, Yainet; OROZCO MONTEAGUDO, Maykel  and  TABOADA CRISPI, Alberto. Detection of Regions of Interest in Images of the Papanicolaou Test. Rev cuba cienc informat [online]. 2017, vol.11, n.1, pp.182-194. ISSN 2227-1899.

ABSTRACT The detection of abnormalities in medical images is used in the classification and detection of abnormalities. The purpose of an anomaly detector is to identify the differences in a series of data, without having any previous information of its properties. The objective of this work is the construction of an anomaly detector for images of the Papanicolaou test, for which an algorithm that can determine regions of interest was designed and implemented. This algorithm was tested with 40 images, 20 containing only normal cells and 20 with abnormal cells. 100% of the images with anomalous cells presented regions of interest. Of the 20 images with only normal cells, only 9 contained regions of interest. On the other hand, the proposed method included 92.43% of the abnormal cells in the regions of interest.

Keywords : Papanicolaou test; detection of anomalies; regions of interest; classification; classification maps.

        · abstract in Spanish     · text in Spanish     · Spanish ( pdf )