Automatic Lumen Segmentation in Intravascular Optical Coherence Tomography Using Morphological Features

Abstract

Lumen segmentation in intravascular optical coherence tomography (IVOCT) images is a fundamental work for more advanced plaque analysis, stent recognition, fractional flow reserve (FFR) assessment, and so on. However, the catheter, guide-wire, inadequate blood clearance, and other factors will impact on the accuracy of lumen segmentation. We present a simple and effective method for automatic lumen segmentation method in IVOCT based on morphological features. We use image enhancement, median filtering, image binarization, and morphological closing operation to reduce speckle noise, minimize the effect of blood artifacts and fill in small holes inside vascular walls. We extract the orientation and area-size of connected regions as morphological features in images and remove the catheter and guidewire completely by morphological corrosion operation, small area-size region removal, and orientation morphological feature comparison, and then the contour of the lumen can be discriminated. The evaluation metrics of this method, the Dice index, Hausdorff distance, Jaccard index, and accuracy of 99.32%, 0.06 mm, 99.4%, and 99.66%, respectively, are obtained from comparing with expert annotations on 268 IVOCT images. Compared with the other morphology-based lumen segmentation methods, the presented method can remove the catheter and guide-wire completely, even if the catheter and guide-wire cling to the lumen or the shape of the catheter is irregular. Since only morphological operations are used to complete all processes, the calculation burden is reduced greatly.

Automated accurate lumen segmentation using L-mode interpolation for three-dimensional intravascular optical coherence tomography

“…The correlation and OCT-derived FFR of lumen extracted using the proposed and alternate schemes were compared. Numerous algorithms for lumen segmentation and 3D reconstruction have been reported [10][11][12][13][14][15][16][17][18][19][20][21] that fail to accurately delineate lumen contours with wide intimal discontinuities. By utilizing the proposed scheme, reliable 3D models can be constructed for OCT-derived FFR computation.…”

“…A key flaw common among the aforementioned algorithms and most others [19][20][21]is the failure to automatically detect lumen contours in images with wide intimal discontinuities due to shadows from guide wires, motion artifacts, or bifurcations. Usually, images with side branches and bifurcations are discarded, limiting the scope of assessment of the mother vessel lumen.…”

Three‐dimensional spatial reconstruction of coronary arteries based on fusion of intravascular optical coherence tomography and coronary angiography

“…The algorithm flow chart is shown in Figure 1. Here lumen segmentation of IVOCT images is completed by a morphological feature method, which has been illustrated in [25]. We will present other steps in the following sections.…”

Dual-Stage U-Shape Convolutional Network for Esophageal Tissue Segmentation in OCT Images

“…We use the following metrics to evaluate the proposed D-UCN framework, including the precision, recall, Dice coefficient, the Hausdorff distance (HD) and the average distance (AVD) [36], [37]. The precision, recall and Dice coefficient evaluated the segmentation performance based on the overlap area, which are defined as Eqs.…”