Pattern recognition in color images: micro-structural analysis of bone tissue around anodized zirconium implants

ABRAS GUILLERMO; MARIA R. KATUNAR; JOSEFINA BALLARRE; VIRGINIA BALLARIN; JUAN I. PASTORE

Cordoba

Congreso; XXI Congreso Argentino de Bioingeniería y X Jornadas de Ingeniería Clínica - SABI 2017; 2017

Organizado por la Sociedad Argentina de Bioingeniería y la Facultad de Ciencias Exactas, Físicas y Naturales y la Facultad de Ciencias Médicas de la Universidad Nacional de Córdoba,

Characterization of newly formed bone around surface-modified Zr implants is one way to determine and quantify the bioactive response of these generated surfaces. We present a clustering algorithm based on weight K-means using different color spaces models for the segmentation of trabecular bone tissue in histological images. The images are from tissue around implanted modified Zr implants. The segmentation of these kinds of images is a complex process since they present diffuse areas. The proposed method was validated using the precision index from ?goal standard? images generated by a group of experts.Keywords? Zirconium (Zr), Images, k-medias, Pattern Recognition, Bone Tissue.