Services on Demand
Journal
Article
Indicators
- Cited by SciELO
- Access statistics
Related links
- Cited by Google
- Similars in SciELO
- Similars in Google
Share
Revista U.D.C.A Actualidad & Divulgación Científica
Print version ISSN 0123-4226
Abstract
RODRIGUEZ, Javier; SANCHEZ, Miguel; BARRIOS, Freddy and SORACIPA, Yolanda. EUCLIDIAN AND FRACTAL GEOMETRY APPLIED TO THE DIAGNOSIS OF LESIONS FROM CERVICAL CELLS. rev.udcaactual.divulg.cient. [online]. 2017, vol.20, n.1, pp.13-21. ISSN 0123-4226.
A diagnostic methodology for preneoplastic and neoplastic lesions of cervical cells from simultaneous fractals and euclidean measures was developed. In this work the objective was to confirm the diagnostic concordance of the methodology in normal cells and in different stages of progression to disease. For this, 60 photographs of cervical squamous epithelium cells were taken: 10 normal, 10 ASCUS, 20 with low grade - squamous intraepithelial lesion (L-SIL) and 20 high-grade lesion (H-SIL). Measures of fractal dimension and space occupancy of the surface and the edge of the nucleus and cytoplasm were performed in Box Counting Fractal space, establishing their physical-mathematical diagnosis. Measurements of the nucleus surface for normality were between 305 and 651, for ASCUS between 1293 and 4588, for L-SIL, between 986 and 4873 and for H-SIL between 567 and 2311. The subtraction of Cytoplasm-Nucleus border was found between 238 and 477 for normality; between 185 and 417 for ASCUS; between 131 and 342 for L-SIL and between 43 and 117 for H-SIL, finding sensitivity and specificity of 100%, likelihood ratio of 0 and kappa coefficient of 1. Conclusions: the diagnostic concordance at clinical level of the physical-mathematical method was confirmed, objective and reproducible quantifying the degree of injury of cervical cells and establishing an objective diagnosis for ASCUS cells from simultaneous fractals and euclidean measures, which improves qualitative classification methods.
Keywords : ASCUS; L-SIL; H-SIL; cervix cancer; cytology.