ENDOBRONCHIAL TUMOR MASS INDICATION IN VIDEOBRONCHOSCOPY - Block based Analysis

摘要

Computer-assisted interpretation of bronchial neoplastic lesion is an innovative but exceptionally challenging task due to highly diversified pathology appearance, video quality limitations and the role of subjective assessment of the endobronchial images. This work is focused on various manifestations of endobronchial tumors in acquired image sequences, bronchoscope navigation, artifacts, lightening and reflections, changing color dominants and unstable focus conditions. Proposed method of neoplasmatic areas indication was based on three steps of video analysis: a) informative frame selection, b) block-based unsupervised determining of enlarged textual activity, c) recognition of potentially tumor tissue, based on feature selection in different domains of transformed image and Support Vector Machine (SVM) classification. Prior to all of these procedures, wavelet-based image processing was applied to extract texture image for further analysis. Proposed method was verified with a reference image dataset containing diversified endobronchial tumor patterns. Obtained results reveal high accuracy for independent classification of individual (single video record) forms of endobronchial tumor patterns. The overall accuracy for whole dataset of 888 test blocks reached 100%. Less complex (approximately two times) procedure including initial blocks of interests selection reached accuracy of 96%.