Classifying cervix tissue patterns with texture analysis

This paper presents a generalized statistical texture analysis technique for characterizing and recognizing typical, diagnostically most important, vascular patterns relating to cervical lesions from colposcopic images. Recognizing the fact that the texture patterns related to cervical lesions are primarily due to the vascular structures, the technique first extracts the vascular structures from the original cervical images, followed by vectorizing the extracted vascular structures with piecewise connecting line segments. Statistical distributions of the line segments are then constructed. First and second order statistics derived from the distributions are used as texture measures for cervical lesion classification. Experimental study demonstrated that the developed algorithm is very promising in discriminating between cervical texture patterns indicative of different stages of cervical lesions. Classification of cervical lesions consisting of six different vascular patterns yielded an average of 95+ percent classification accuracy. The major contributions of this research include development of a novel generalized statistical pattern recognition approach for accurately characterizing cervical textures and defining a set of textural features that capture specific characteristics of the cervical textures as perceived by human. These contributions lead to a system that can recognize typical vascular patterns indicative of different stages of cervix lesions.