PROJECT TITLE :

Pulmonary fissure detection in ct images using a Derivative of stick filter - 2016

ABSTRACT:

Pulmonary fissures are vital landmarks for recognition of lung anatomy. In CT pictures, automatic detection of fissures is complicated by factors like intensity variability, pathological deformation and imaging noise. To circumvent this drawback, we have a tendency to propose a derivative of stick (DoS) filter for fissure enhancement and a post-processing pipeline for subsequent segmentation. Considering a typical skinny curvilinear form of fissure profiles inside 2D cross-sections, the DoS filter is presented by 1st defining nonlinear derivatives along a triple stick kernel in varying directions. Then, to accommodate pathological abnormality and orientational deviation, a max-min cascading and multiple plane integration theme is adopted to form a shape-tuned probability for 3D surface patches discrimination. During the post-processing stage, our main contribution is to isolate the fissure patches from adhering clutters by introducing a branch-point removal algorithm, and a multi-threshold merging framework is utilized to atone for native intensity inhomogeneity. The performance of our methodology was validated in experiments with two clinical CT knowledge sets together with fifty five publicly available LOLA1one scans in addition to separate left and right lung images from 23 GLUCOLD scans of COPD patients. Compared with manually delineating interlobar boundary references, our methodology obtained a high segmentation accuracy with median F1-scores of zero.83three, zero.885, and zero.856 for the LOLA1one, left and right lung pictures respectively, whereas the corresponding indices for a conventional Wiemker filtering technique were zero.687, 0.853, and zero.841. The smart performance of our proposed method was additionally verified by visual inspection and demonstration on abnormal and pathological cases, where typical deformations were robustly detected along with traditional fissures.