Interval‐coded texture features for artifact rejection in automated cervical cytology

• Published in: Cytometry (New York, N.Y.) Vol. 9; no. 5; pp. 418 - 425
• Main Authors: Tucker, James H., Rodenacker, Karsten, Juetting, Uta, Nickolls, Peter, Watts, Keith, Burger, Georg
• Format: Journal Article
• Language: English
• Published: Hoboken Wiley Subscription Services, Inc., A Wiley Company 01-09-1988; Wiley-Liss
• Subjects: automated prescreening; Biological and medical sciences; Cell Division; cervical cytology chromatin; Cytological Techniques; Diagnosis, Computer-Assisted - methods; Diverse techniques; DNA, Neoplasm - analysis; Female; Fundamental and applied biological sciences. Psychology; Humans; Image analysis; Image Processing, Computer-Assisted - methods; Molecular and cellular biology; pattern recognition features; texture; Uterine Cervical Neoplasms - analysis; Uterine Cervical Neoplasms - pathology; Image analysis; Chromatin; Artefact; Automated cervical cytology prescreener; Cytology
• ISSN: 0196-4763; 1097-0320
• DOI: 10.1002/cyto.990090503

In order to improve the separation between abnormal cells and noncellular artifacts in the CERVIFIP automated cervical cytology prescreening system, 22 different object texture features were investigated. The features were all statistical parameters of the pixel density histograms or one‐dimensional filtered values of central and border regions of the object images. The features were calculated for 231 images (100 cells and 131 artifacts) detected as Suspect Cells by the current CERVIFIP and were then tested in hierarchical and linear discriminant classifiers. After selecting the two best features for use in a hierarchical classifier, 83% correct classification was achieved. One of these features was specifically designed to remove poorly focused objects. With maximum likelihood discrimination using all 22 features, an overall correct classification rate of 90% was obtained.