Fast and Accurate Semiautomatic Segmentation of Individual Teeth from Dental CT Images

Abstract

DIn this paper, we propose a fast and accurate semiautomatic method to effectively distinguish individual teeth from the sockets of teeth in dental CT images. Parameter values of thresholding and shapes of the teeth are propagated to the neighboring slice, based on the separated teeth from reference images. After the propagation of threshold values and shapes of the teeth, the histogram of the current slice was analyzed. The individual teeth are automatically separated and segmented by using seeded region growing. Then, the newly generated separation information is iteratively propagated to the neighboring slice. Our method was validated by ten sets of dental CT scans, and the results were compared with the manually segmented result and conventional methods. The average error of absolute value of volume measurement was 2.29 ± 0.56%, which was more accurate than conventional methods. Boosting up the speed with the multicore processors was shown to be 2.4 times faster than a single core processor. The proposed method identified the individual teeth accurately, demonstrating that it can give dentists substantial assistance during dental surgery.

Automated CNN-Based Tooth Segmentation in Cone-Beam CT for Dental Implant Planning

“…Therefore, teeth and their surrounding structures cannot easily be separated by a simple thresholding method [12]. To address this problem, some approaches have utilized adaptive thresholding methods [13], [14]. However, these approaches have some limitations since they require manual selection of the reference slice to select the initial contours.…”

Convolutional neural network–based metal and streak artifacts reduction in dental CT images with sparse‐view sampling scheme

“…Moreover, it can also be improved by using more advanced metal object segmentation methods. [44][45][46] FSNMAR method 47 has been proposed to compensate the sharpness of teeth or soft tissue around the metal objects that can be lost after NMAR processing. It is effective when the density of the metal objects is not significantly large and the information of other teeth or soft tissue still remains around the metal objects.…”

Is Manual Segmentation the Real Gold Standard for Tooth Segmentation? A Preliminary <i>in vivo</i> Study Using Cone-beam Computed Tomography Images

“…Recently, several segmentation methods and processing algorithms are being actively developed to overcome these problems. Some have proposed to segment the crown and the root separately with two level set approaches (3,4,12,28) and others have proposed improved hybrid active contour model to accurately distinguish tooth structure from its surroundings (9,39). However, in most of these studies, segmentation accuracy was tested by methods that could only be implemented under in vitro conditions (9,12,28,39).…”

“…The determination of the optimum threshold can be formulated as a minimization problem of histogram curve of the image and can be estimated by finding the local minima between the peaks points corresponding to the teeth and the background in the histogram. In this method, the optimum threshold value must be the largest threshold value that prevents over segmentation while maintaining the shape and size of the segmented teeth in the previous slice (39). On each slice of CBCT images of 10 teeth, manual (Figure 2a), semi-automatic (Figure 2b) and automatic (Figure 2c) segmentation were performed.…”