A robust fully automatic lumen segmentation method for in vivo intracoronary optical coherence tomography

Abstract

Introduction: Intravascular optical coherence tomography (IVOCT) is an in-vivo imaging modality based on the introduction of a catheter in a blood vessel for viewing its inner wall using electromagnetic radiation. One of the most developed automatic applications for this modality is the lumen area segmentation, however on the evaluation of these methods, the slices inside bifurcation regions, or with the presence of complex atherosclerotic plaques and dissections are usually discarded. This paper describes a fully-automatic method for computing the lumen area in IVOCT images where the set of slices includes complex atherosclerotic plaques and dissections. Methods: The proposed lumen segmentation method is divided into two steps: preprocessing, including the removal of artifacts and the second step comprises a lumen detection using morphological operations. In addition, it is proposed an approach to delimit the lumen area for slices inside bifurcation region, considering only the main branch. Results: Evaluation of the automatic lumen segmentation used manual segmentations as a reference, it was performed on 1328 human IVOCT images, presenting a mean difference in lumen area and Dice metrics of 0.19 mm 2 and 97% for slices outside the bifurcation, 1.2 mm 2 and 88% in the regions with bifurcation without automatic contour correction and 0.52 mm 2 and 90% inside bifurcation region with automatic contour correction. Conclusion: This present study shows a robust lumen segmentation method for vessel cross-sections with dissections and complex plaque and bifurcation avoiding the exclusion of such regions from the dataset analysis.

A novel automated lumen segmentation and classification algorithm for detection of irregular protrusion after stents deployment

“…In the present study, the results of JSC, DSC, MADA, HD, RMSD, and accuracy will now be compared with the findings of previous work. The current study found the average of JSC as 95.4%, 95.5%, and 94.9% for total score, normal, and abnormal cases, respectively, whereas Tsantis et al, Macedo et al, and Miyagawa et al obtained JSC as 93.8%, 94.7%, and 97.6%, respectively. Wang et al, Tung et al, Ughi et al, Macedo et al, Cao et al, and Miyagawa et al calculated the average value of DSC as 97.0%, 97.0%, 97.0%, 97.3%, 98.1%, and 98.8% respectively, while our method reached an average value the DSC as 97.6%, 97.6%, and 97.2% for total score, normal, and abnormal cases, respectively.…”

“…Several researchers have reported automatic segmentation using old algorithms. Macedo et al used an automatic segmentation method that was divided into two steps: preprocessing, containing the removal of artifacts and the second step comprises a lumen detection using morphological operations. Athanasiou et al used an old algorithm, three‐dimensional reconstruction, for the lumen segmentation, which was based on bilateral filtering and K‐mean algorithm.…”

“…The labeled dataset of the IVOCT images was based on the clinical reading result per IVOCT frame. We built the dataset of 14 207 images from 54 patients who receive drug‐eluting stent, and lumen segmentation from 1541 images was manually performed to validate performance of our lumen segmentation algorithm. Compared with previous studies, the dataset is a large dataset to demonstrate the validity and accuracy of this study.…”

Reconstruction of stented coronary arteries from optical coherence tomography images: Feasibility, validation, and repeatability of a segmentation method

“…Over the last years, several methods have been proposed to analyze OCT datasets in order to automatically detect the lumen contours [8–22] and stent struts [9,10,16–18,20,22–27]. In general, those automatic algorithms showed good results with respect to manual analysis.…”

Bimodal intravascular volumetric imaging combining OCT and MPI

“…5(b) is used. The boundary is segmented with an edge detection algorithm after several preprocessing steps (removal of catheter, filtering, thresholding, morphological operations) as introduced by de Macedo et al 33 After lumen segmentation, we need to spatially align the acquired sequence of Cartesian B-scans and containing lumen areas. With the pullback distance s = 30 mm and distance l = 25 mm of the phantom landmarks, each IVOCT pullback consists of B-scans with narrowed lumen diameter.…”