Open AccessClassification of Optical Images of Cervical Lymph Node Cells
Author(s) -
Salim J. Attia
Publication year - 2020
Publication title -
iop conference series. materials science and engineering
Language(s) - English
Resource type - Journals
eISSN - 1757-899X
pISSN - 1757-8981
DOI - 10.1088/1757-899x/928/7/072049
Subject(s) - standard deviation , perimeter , pixel , pattern recognition (psychology) , segmentation , artificial intelligence , computer science , cervical lymph nodes , euclidean distance , mode (computer interface) , mathematics , statistics , medicine , geometry , cancer , metastasis , operating system
the study considers the optical classification of cervical nodal lymph cells and is based on research into the development of a Computer Aid Diagnosis (CAD) to detect the malignancy cases of diseases. We consider 2 sets of features one of them is the statistical features; included Mode, Median, Mean, Standard Deviation and Maximum Probability Density and the second set are the features that consist of Euclidian geometrical features like the Object Perimeter, Area and Infill Coefficient. The segmentation method is based on following up the cell and its background regions as ranges in the minimum-maximum of pixel values. The decision making approach is based on applying of Minimum Distance which give accuracy of 97%.