Accurate and reliable segmentation of liver tissue and liver tumor is essential for the follow-up of hepatic diagnosis. In this paper, we present a method for liver segmentation and a method for liver tumor segmentation. The two methods are grounded on a novel unified level set method (LSM), which incorporates both region information and edge information to evolve the contour. This level set framework is more resistant to edge leakage than the single-information driven LSMs for liver segmentation and surpasses many other models for liver tumor segmentation. Specifically, for liver segmentation, a hybrid image preprocessing scheme is used first to convert an input CT image into a binary image. Then with manual setting of a few seed points on the obtained binary image, the following region-growing is performed to extract a rough liver region with no leakage. The unified LSM is proposed at last to refine the segmentation result. For liver tumor segmentation, a local intensity clustering based LSM coupled with hidden Markov random field and expectation-maximization (HMRF-EM) algorithm is applied to construct an enhanced edge indicator for the unified LSM. With this development, expected segmentation results can be obtained via the unified LSM, even for complex tumors. The two methods were evaluated with various datasets containing a local hospital dataset, the public datasets SLIVER07, 3Dircadb, and MIDAS via five measures. The proposed liver segmentation method outperformed other previous semiautomatic methods on the SLIVER07 dataset and required less interaction. The proposed liver tumor segmentation method was also competitive with other state-of-the-art methods in both accuracy and efficiency on the 3Dircadb database. Our methods are evaluated to be accurate and efficient, which allows their adoptions in clinical practice.
Segmentation is an image processing operation for identifying an anatomical structure from the surrounding tissues. In the area of computed tomography (CT) based clinical hepatic diagnosis, accurate and reliable segmentation of liver tissue and liver tumor is essential for the follow-up treatment planning and evaluation and computer-aided surgery. In current clinical practice, manual delineation of liver and liver tumor on each slice is still typically performed by radiologists, which could obtain the arguably most accurate segmentation results, but is time-consuming, tedious, and laborious and introduces interobserver variability. Additionally, due to the blurry edges and low level of contrast characterizing the CT images, liver segmentation is regarded as a challenging task. The segmentation of liver tumor encounters the same problem due to the low contrast, ambiguous boundaries, and intensity inhomogeneities. Therefore, the development of sophisticated segmentation algorithms has become a major research focus in medical image computing with the potential to provide accurate, effective, and robust approaches for clinical practice. In the past decade, many remarkable liver and liver tumor segmentation methods have been presented with varying degrees of success. These methods can be roughly classified into two categories: automatic and semiautomatic methods, depending on whether the user interaction is required.
Among automatic liver segmentation algorithms, model-based methods have proved to be the most effective one, where prior anatomical knowledge of the target organ is incorporated into the segmentation process [
An increasing number of automatic methods are already available for liver tumor segmentation. For instance, Goetz et al. [
However, fully automatic methods for liver and liver tumor segmentation, such as aforementioned segmentation approaches, sometimes require massive training datasets, bringing time-consuming training process and statistical model construction. Moreover, they generally suffer from lower accuracy and robustness, as well as a significant higher computational cost [
For semiautomatic liver segmentation, Yang et al. [
For semiautomatic liver tumor segmentation, Häme et al. [
One of the reasons that discourage the use of semiautomatic segmentation approaches is that high interaction is required. For instance, among some aforementioned approaches, the type of high interaction applied varies from setting multiple seed points to extensive manual refinement for postprocessing. Moreover, the segmentation accuracy still leaves a room for improvement. Thus, in order to achieve more accurate and efficient segmentation with less interaction, we present a reliable framework for liver and liver tumor segmentation based on a novel unified LSM. In the following, we briefly introduce the LSMs and explain how previous work differs from ours.
Techniques based on LSMs have been extensively investigated for liver and liver tumor segmentation [
Existing LSMs can be categorized into two major classes: region-based models and edge-based models. However, both of them have inherent limitations for liver and liver tumor segmentation: (1) Considering liver segmentation, most used LSMs are edge-based models that introduce a speed function such as the balloon force term to shrink or expand the contour, since region-based models mostly are unable to detect the objective liver boundaries accurately. However, due to the existence of nonliver tissues such as vessels and tumors within the liver, edge-based LSMs with a small balloon force or insufficient iterations may not pass through the nonliver tissues, leading to undersegmentation. In contrast, if the balloon force is large or the iterations are excessive, the contours will pass through weak liver boundaries, leading to oversegmentation. So it is difficult to decide an appropriate group of the balloon force and evolution iterations for edge-based LSMs. (2) For liver tumor segmentation, the performance of edge-based LSMs relies on the precondition that liver tumors have clear and distinct edges. Unfortunately, liver tumors are often ambiguous, and edge-based LSMs will fail to identify the edges. On the other hand, although region-based LSMs are applicable to liver tumor segmentation, they have an inherent disadvantage that they are not competent to tackle complex liver tumors with low contrast and intensity inhomogeneities [
Rather than modifying or enhancing specific terms in the edge-based LSM or the region-based LSM to propose an improved LSM, we construct a unified level set framework by incorporating both edge information and region information of image to control the contour evolution. This framework is the core of the proposed methods for liver and liver tumor segmentation in this paper. Specifically, our proposed liver segmentation method mainly consists of three stages, a hybrid image preprocessing scheme to transform an original CT image into a binary image, region-growing to initially extract a rough liver region, and the unified LSM to refine the initial liver segmentation; our proposed liver tumor segmentation method mainly consists of three components, a region-based LSM proposed by Li et al. [
Generally, LSMs can be applied to 2D segmentation by evolving curves, as well as 3D segmentation by evolving surfaces. The 3D level set method is theoretically a very nice solution, for it could produce a smoother boundary in the axial direction, leading to better incorporation of the 3D geometry. However, due to the complicated CT image conditions such as the noise and neighboring objects, the 3D segmentation is very sensitive to the initialization and suffers from the convergence to the local minima [
The anisotropy of abdominal CT data is mainly reflected in the various organs and tissues with complex anatomical structures and different intensities and the irregularly distributed noise within image. Our algorithms are able to resist and deal with the anisotropy of CT data, as well as the blurred edges, low contrast, and intensity inhomogeneities characterizing the CT data. In summary, the main contributions of this paper are as follows: We propose a new unified LSM. This framework is more resistant to edge leakage than the single-information driven LSMs for liver segmentation and surpasses many other models for liver tumor segmentation. It is able to obtain more accurate results for liver and liver tumor segmentation. We propose a hybrid image preprocessing scheme to convert the original CT image into a binary image. With this conversion, the median number of seed points required per CT image for region-growing to initially extract a rough liver region is 1 (range, 1 to 8). Besides, threshold setting and initial seed location setting become simpler, and there is no risk of oversegmentation. To tackle more complex liver tumors, we construct an enhanced edge indicator for the unified LSM during liver tumor segmentation. With this development, our LSM can better handle the tumors with low contrast, ambiguous edges, and intensity inhomogeneities. We extensively validate our methods with various datasets to show their accuracy and efficiency. The liver segmentation method outperforms other previous semiautomatic methods and requires less interaction on the SLIVER07 dataset. The liver tumor segmentation method is also competitive with other state-of-the-art methods on the 3Dircadb dataset.
The rest of this paper is organized as follows. A complete methodology of the proposed methods for liver and liver tumor segmentation is elaborated in Section
An overview of the proposed methods for liver and liver tumor segmentation is illustrated in Figure For liver segmentation, a hybrid image preprocessing scheme consisting of an anisotropic filter [ For liver tumor segmentation, the user is required to define a tumor ROI manually first, where the following segmentation process will perform. Then the region-based LSM and the HMRF-EM algorithm are applied to construct an enhanced edge indicator for the unified LSM, which is performed at last to obtain the segmentation result.
Flowchart of the proposed methods for liver and liver tumor segmentation. The left corresponds to liver segmentation, and the right corresponds to liver tumor segmentation.
In the following, we give a detailed illustration of preprocessing. Instances of the results obtained by the hybrid image preprocessing are shown in Figure
Results of hybrid image preprocessing: (a) original image; (b) after anisotropic filter; (c) after scale-specific gradient magnitude filter; (d) after nonlinear grayscale conversion; (e)-(h), respectively, indicate the results of binary conversion with
Given an original image
The scale-specific gradient magnitude filter is employed on the noise-reduced CT image
A nonlinear grayscale converter based on a sigmoid function is applied to enhance the contrast of the liver parenchyma, as illustrated in Figure
We propose a customized binarization method as the final preprocessing step to convert the image
The role of proposed binarization method is reflected in the following two aspects: During Step 2, the scale-specific gradient magnitude filter not only enhances the liver boundaries, but also inevitably strengthens the extra noise within the liver. Binary conversion proposed here can reduce the noise enhancement. Results of noise reduction with binary conversion under different values of In our study, conventional region-growing is used. Namely, a number of seed points are manually selected by users, and the initial region begins as the exact locations of these seed points. The regions are then grown from these seed points to adjacent pixel points depending on a homogeneity criterion. In our study, we define the homogeneity criterion by gray value. Specifically, a gray-level threshold
To clearly demonstrate the difference between region-growing for original CT image and region-growing for binary image, a comparison of them was made. We randomly selected two original CT images, defined them as image M and image N, and normalized their gray values to the range of 0 to 1. As illustrated in Figure
The comparison between region-growing for original CT image and region-growing for binary image. (a)-(b) indicate the original image M, (e)-(f) indicate the binary image of image M, (g)-(j) indicate the original CT image N, and (k)-(l) denote the binary image of image N. And images below (a)-(l) denote the corresponding results of region-growing. The initial seed points are denoted by red points, and
In this section, region-growing is performed on the obtained binary image to initially extract a rough liver region. The hybrid image preprocessing scheme provides a good condition for seed growth, as illustrated in Figure
Four basic cases of seed settings and corresponding initial segmentation results: (a) a liver having an indiscrete region; (b) a liver having two discrete regions; (c) a liver having an inside vessel; (d) a liver having an inside liver tumor.
Note that region-growing for a liver in a more complex case (such as a liver having discrete regions and inside tumors) may require more seed points to ensure that all the liver areas could be segmented. According to statistics, in our liver segmentation experiment, the median number of seed points required per CT image for region-growing to initially extract a rough liver region is 1 (range, 1 to 8), and more details can be found in Section
From Figure
The region-based LSM is not preferred for liver segmentation as it is often unable to detect the objective liver boundaries correctly. And the edge-based LSM encounters the difficulty of setting an appropriate group of the balloon force and iterations. In order to tackle these limitations, we propose a unified double-information driven LSM for liver segmentation. With an enhanced edge indicator, this level set framework is applicable for complex liver tumor segmentation as well. Compared with other two single-information driven LSMs, our LSM has proved to be able to adapt to a larger balloon force and more iterations (the comparisons can be found in Section
Let
We define an energy functional
We use an original edge indicator function
Then
Zhang et al. [
For the given image
With the LSM, we assume
By minimizing
We integrate the SPF to
Thus, the unified LSM could be expressed as
As for the stop criterion of the LSM refinement for liver segmentation, on the one hand, the contour propagation will stop if the maximum number of iterations is reached. On the other hand, the user is allowed to terminate the LSM evolution manually. This additional manual termination is proposed to ensure the unified LSM could bring expected segmentation even in extreme cases, such as the case of almost complete absence of edge between the liver and adjacent tissues, where the LSM is prone to cause leakage due to lack of edge information. Note that this additional operation has no significant effect on the interaction burden since the proportion of extreme cases is small (a detailed interaction analysis can be found in Section
Examples of the unified LSM initialization and refinement are illustrated in Figure
Examples of the unified LSM initialization and refinement: (a) binary image obtained from region-growing; (b) LSM initialization; (c) the result of refinement.
The purpose of a ROI definition is to limit the tumor segmentation to the selected region. Advantages of the ROI definition are as follows:
An example of a ROI definition.
Original edge indicator
Assume the observed image
Derive a local intensity clustering property. Let
Apply the standard K-means clustering to classify the local intensities and derive a level set energy formulation. The domain of the image is segmented into two disjoint regions
By using the DRLSE frame and adding the energy
Due to the robustness of this LSM to contour initialization, we use a random initialization mechanism to initialize the contour automatically, avoiding manual labor [
An example of a segmentation result obtained through this LSM is shown in Figure
Results of the region-based LSM and the HMRF-EM scheme: (a) an original ROI; (b) the region-based LSM segmentation; (c) the binary mask of (b); (d) the improved binary classification obtained via the HMRF-EM algorithm.
Given an image
And
Assume
With these assumptions and conditions mentioned above, the HMRF-EM algorithm used in this paper could be described as follows, and the binary mask provides the initial labels
(1) Start with the initial condition provided by the binary mask.
(2) Calculate the likelihood distribution
(3) Estimate the class labels with current
(4) Calculate the posterior distribution for all pixel intensities
(5) Update parameters with
(6)
Through the HMRF-EM framework, an improved binary classification having advantages of the two mentioned methods is obtained, as illustrated in Figure
Figure
(a) original
By replacing the original
The unified LSM for liver tumor segmentation requires manual initialization, and the initial contour is preferred to be located inside the liver tumor since the enhanced edge indicator allows faster contour propagation in white area. In our study, we initialize the LSM with a rectangle inside the liver tumor. Additionally, unlike the liver segmentation, no additional interaction is required to terminate the level set evolution due to the enhanced edge indicator, whereas contour propagation will stop when the maximum number of iterations is reached.
Successful segmentation examples of two challenging liver tumors are displayed in Figure
Two examples of challenging liver tumor segmentation: (rows) from top to bottom: a tumor case with an ambiguous and variable edge and a tumor case with low contrast and intensity inhomogeneities; (columns) from left to right: the original ROIs, the enhanced edge indicators, the segmentation results denoted by red lines, and the results shown in full images.
The proposed methods were implemented in C++/mex and Matlab environment on a Windows-based computer with an i5-2400 3.1GHz CPU, AMD Radeon HD 6450 GPU, and 6GB RAM. Details of the used validation data are given below: For liver segmentation, the applied data came from the SLIVER07-Train database and the 3Dircadb database. The SLIVER07-Train database contains 20 contrast-enhanced CT volumes with standard segmentation. All the volumes have an in-plane resolution of Three datasets were used for liver tumor segmentation method validation. They are a local dataset acquired at Ningbo Li Hui-Li hospital, China, the MIDAS dataset provided by the Imaging Science and Information Systems (ISIS), and the 3Dircadb dataset. There are 10 liver tumors in the hospital data coming from 3 patients; each image has a matrix size of
For accuracy evaluation, five evaluation measures were used to compare each segmentation result with its corresponding reference segmentation. They are volumetric overlap error (VOE, %), relative absolute volume difference (RVD, %), average symmetric surface distance (ASD, mm), root mean square symmetric surface distance (RMSD, mm), and maximum symmetric surface distance (MSD, mm) [
Here, we give the values of the most important parameters. We set
9 groups of the balloon force and iterations and corresponding RVD values.
Groups | RVD/% |
---|---|
| -6.18 |
| -2.02 |
| 0.85 |
| -4.17 |
| 1.05 |
| 4.67 |
| -2.69 |
| 3.37 |
| 8.34 |
In this section, to demonstrate the superiority of the proposed unified LSM for liver segmentation, we made three comparisons of the unified LSM and other two single-information driven LSMs. The three LSMs are all in the DRLSE framework, respectively, represented by
(1)
(2)
(3)
First comparison: (rows) from top to bottom: the results of the edge-DRLSE, the region-DRLSE, and the unified-DRLSE, respectively; (columns): (a) level set initialization; (b) results under iteration 50; (c) results under iteration 100; (d) results under iteration 150.
From the results, we can note that, for the edge-DRLSE, there is already an obvious leakage when the number of iterations reaches 50. As the number of iterations increases, the leakage becomes more and more serious. The region-DRLSE is unable to detect the liver boundary correctly, leading to global segmentation. In contrast, as for the unified-DRLSE that is under a larger balloon force, we can observe that its leakage changes imperceptibly with the increase in iterations. This qualitative comparison shows the unified-DRLSE is able to adapt to a larger balloon force and more iterations.
As illustrated in Figure
Second comparison: we set
Image A
Image B
Third comparison: bar chart: three LSMs set with
Note that, in order to remain objective during the three comparisons, no additional manual termination was added; instead, contour evolution stopped when the maximum number of iterations was reached. Results of the above three comparisons were in line with our assumptions. That is, the region-DRLSE often failed to observe the objective liver edges accurately, leading to the worst performance among the three. The edge-DRLSE was able to detect the liver boundaries, but it was unable to stay stable under a larger balloon force or more iterations. So it outperformed the region-DRLSE but was not as good as the unified-DRLSE. The unified LSM achieved the best performance, which proves that combining more image information is able to make the model more stable and lead to more accurate segmentation.
The proposed liver segmentation method was validated with the two public datasets via five measures. Validation results are illustrated in Table
Segmentation performance of the proposed method on the two public datasets.
VOE | RVD | ASD | RMSD | MSD | |
---|---|---|---|---|---|
SLIVER07 | |||||
| 4.86 | -0.70 | 0.90 | 1.59 | 7.82 |
| 3.09 | 3.30 | 0.66 | 1.41 | 7.86 |
| 24.06 | -18.03 | 4.51 | 8.63 | 49.90 |
| 0.37 | 0 | 0.13 | 0.20 | 0.66 |
3Dircadb | |||||
| 6.73 | -1.02 | 1.29 | 2.04 | 9.68 |
| 3.10 | 3.76 | 0.56 | 1.18 | 7.42 |
| 22.62 | -18.33 | 3.65 | 6.95 | 45.03 |
| 1.32 | 0 | 0.21 | 0.37 | 0.78 |
For the SLIVER07 challenge data, independent delineations were provided by the organization of the challenge. Based on the comparison a score was given to each validation measure and to each segmentation result of the challenge data. A score of 100 was assigned to the perfect segmentation, i.e., a value of 0 for each validation measure. Reference values of VOE = 6.4%, RVD = 4.7%, ASD = 1.0 mm, RMSD = 1.8 mm, and MSD=19 mm were assigned with a score of 75. Therefore, a score of 75 meant the segmentation was as good as the manual delineation. The total score was the average of the scores for each validation measure. The scores received from evaluation are illustrated with a boxplot in Figure
Comparison of the proposed method with previous semiautomatic methods on the SLIVER07 dataset.
Method | Interaction | VOE | RVD | ASD | RMSD | MSD | Score |
---|---|---|---|---|---|---|---|
Beichel et al. [ | High | 5.2 | 1.0 | 0.8 | 1.4 | 15.7 | 82 |
Dawant et al. [ | Medium | 7.2 | 2.5 | 1.1 | 1.9 | 17.1 | 76 |
Beck et al. [ | High | 6.6 | 1.8 | 1.0 | 1.9 | 18.5 | 77 |
Chartrand et al. [ | High | 5.1 | 1.2 | 1.0 | 2.1 | 21.3 | N/A |
Eapen et al. [ | N/A | 7.3 | 1.3 | 1.1 | 1.7 | 15.7 | 78 |
Zareei et al. [ | N/A | 1.9 | 2.2 | 1.8 | 2.6 | 10.6 | N/A |
| | | | | | | |
Scores of the SLIVER07 dataset, presented as a boxplot, where squares indicate the mean values. Score of average interobserver variability (75) is shown with dashed line for reference.
As illustrated in Table
Comparison of the proposed method with previous automatic methods on the 3Dircadb dataset.
Method | Time | VOE | RVD | ASD | RMSD | MSD |
---|---|---|---|---|---|---|
Li et al. [ | 1.62 min | 9.1 | -0.1 | 1.6 | 3.2 | 28.2 |
Erdt et al. [ | 0.75 min | 10.3 | 1.6 | 1.7 | 3.51 | 26.8 |
Proposed | 24.6 min | 6.7 | -1.0 | 1.3 | 2.0 | 9.7 |
Some randomly selected liver segmentation results denoted in red are shown in Figure
Some randomly selected liver segmentation results denoted in red: (a)-(d) indicate four simple examples of healthy livers having an indiscrete region; (e)-(h) indicate four examples of livers having discrete regions; (i)-(l) indicate four examples of pathological livers.
Here we also give the values of the most important parameters of the proposed method for liver tumor segmentation. For the region-based LSM, we set
In this section, to demonstrate the superiority of the unified LSM for liver tumor segmentation, we made a qualitative comparison of our method and many other methods. As illustrated in Figure
Comparison among our method and many other methods. The ground truth is denoted by green lines while the results obtained by different methods are in red. Columns from left to right indicate the results of, respectively, (a) the C-V model, (b) the HMRF-EM algorithm, (c) the region-DRLSE driven by the SPF, (d) the edge-DRLSE driven by the original
The proposed liver tumor segmentation method was validated with three datasets via five measures. Validation results are illustrated in Table
Segmentation performance of the proposed liver tumor segmentation method on three datasets.
VOE | RVD | ASD | RMSD | MSD | |
---|---|---|---|---|---|
| |||||
Mean | 15.52 | 9.02 | 1.87 | 2.50 | 7.55 |
SD | 7.01 | 6.96 | 0.97 | 1.18 | 3.71 |
Worst | 38.46 | 26.56 | 6.43 | 7.11 | 17.00 |
Best | 3.91 | 0 | 0.54 | 0.73 | 1.41 |
| |||||
Mean | 32.19 | 10.07 | 1.51 | 1.92 | 4.09 |
SD | 11.13 | 28.00 | 0.41 | 0.45 | 1.43 |
Worst | 86.96 | 93.55 | 2.97 | 3.85 | 10.94 |
Best | 15.29 | 0 | 0.68 | 1.1 | 1.76 |
| |||||
Mean | 28.22 | -8.46 | 1.81 | 2.35 | 5.77 |
SD | 11.90 | 18.45 | 1.28 | 1.70 | 4.61 |
Worst | 64.52 | -63 | 6.77 | 9.03 | 25.64 |
Best | 4.59 | 0 | 0.19 | 0.36 | 0.70 |
For the hospital data, we built the ground truth with a radiologist from Ningbo Li Hui-Li Hospital, China. The hospital data includes 10 tumors, and most tumors have more obvious boundaries and higher contrast, tending to be easily segmented. Our method performed well on this dataset. The quantitative results for this training dataset were 15.52%, 9.02%, 1.87 mm, 2.5 mm, and 7.55 mm, respectively.
The MIDAS dataset provides manual segmentation from 5 radiologists, so it allows us to evaluate the robustness of our method according to different manual delineations. Note that we were only interested in metastasis tumors in this validation experiment, so the tumor from patient 4 was not used for method validation because it is a Hepatocellular carcinoma (HCC) [
Five sets of validation results acquired by reference to corresponding manual delineations from five radiologists.
We finally evaluated the proposed method with the public 3Dircadb data. This dataset provides 121 tumors and it is widely used for liver tumor segmentation method validation [
Comparison of the proposed method with state-of-the-art methods on the 3Dircadb dataset.
Method | Mode | Time | VOE | RVD | ASD | RMSD | MSD |
---|---|---|---|---|---|---|---|
Moghbel et al. [ | Auto | 30s/slice | 22.8 | 8.6 | N/A | N/A | N/A |
Sun et al. [ | Auto | 1s/slice | 15.6 | 5.8 | 2.0 | 2.9 | 7.1 |
Wu et al. [ | Semi | 45s | 29.0 | -2.2 | 0.7 | 1.1 | 4.3 |
Li et al. [ | Semi | N/A | 14.4 | 8.1 | 2.4 | 2.9 | 7.2 |
Foruzan et al. [ | Semi | 154s | 30.6 | 16.0 | 4.2 | 5.1 | 12.6 |
Li et al. [ | Auto | 30s-200s | 11.7 | -0.0 | 0.6 | 1.9 | N/A |
| | | | | | | |
Results of volume metrics (VOE and RVD) for the 3Dircadb data, presented as a boxplot, where squares indicate the mean values.
Results of surface metrics (ASD, RMSD, and MSD) for the 3Dircadb data, presented as a boxplot, where squares indicate the mean values.
Some randomly selected liver tumor segmentation results denoted by red lines are presented in Figure
Some randomly selected liver tumor segmentation results denoted by red lines.
In this section, we discuss the interaction time and running time in our study.
For liver segmentation, there are two parts to the interaction. On the one hand, the users are required to set seed points manually for region-growing. We calculated the average number of seed points needed for each CT image of all 40 CT volumes. As illustrated in Figure
Bar chart: average number of seed points required for each CT image of all 40 CT volumes.
For liver tumor segmentation, the users are required to define the ROI and initialize the unified LSM manually. The average total interaction time was about 26 s per tumor according to statistics.
For liver segmentation, plus the interaction time, the average running time needed for each CT sequence was about 25 min, most of which was spent by the evolution of the unified LSM.
For liver tumor segmentation, plus the interaction time, the average running time needed was about 162 s per tumor. Specifically, time for the region-based LSM evolution only took up a small proportion while the HMRF-EM process and the unified LSM propagation spent the most running time. Running time of our method is modest, as illustrated in Table
Figure
Two randomly selected surface renderings of liver and two randomly selected surface renderings of liver tumor. The left columns indicate the 3D visualizations of the ground truth, the middle columns indicate the 3D visualizations obtained via the proposed methods, and the right columns denote the surface distances between the first two, where red (positive value) indicates that the surfaces of the middle models are situated outside the surfaces of the left models and blue (negative value) inside.
In this paper, we have, respectively, presented a method for liver segmentation and a method for liver tumor segmentation. The two methods are grounded on a novel unified LSM, which is driven by both edge information and region information of image. In the following, we will discuss the advantages and shortcomings of the two methods, respectively.
The proposed liver segmentation method consists of a hybrid image preprocessing scheme, region-growing, and the unified LSM. The hybrid image preprocessing converts the input CT image into a binary image, providing a good condition for seed growth. In this way, threshold setting and seed initial location setting become simple and reliable. Moreover, region-growing requires fewer seed points to extract a rough liver region without risk of leakage. The unified LSM is applied at last for refinement to achieve optimal segmentation. Considering liver segmentation, three comparisons of our LSM and other two single-information driven LSMs have proved our LSM is able to adapt to a larger balloon force and more iterations, leading to more accurate segmentation results. The three comparisons are a qualitative comparison, a quantitative comparison, and a statistical comparison. From the result of qualitative comparison, we can observe visually that the proposed LSM is more resistant to edge leakage. In the quantitative comparison, VOE and RVD were used. From the result we can see that, with the increase in iterations, the curves of VOE and RVD belonging to the unified LSM are always under those belonging to other two LSMs and are more flat, indicating that the proposed LSM is able to yield more accurate results as well. The statistical comparison conducted at last makes the first two conclusions more convincing.
Method for liver segmentation was validated with two popular public datasets. Our method surpassed other previous semiautomatic methods on the SLIVER07 datasets, and all measures had a score above 75. Besides, the proposed method was compared with other two automatic methods on the 3Dircadb dataset. Though the two automatic methods were more time-efficient, the segmentation accuracy was lower than that of our method. Thus, we can conclude that the proposed method is competitive with both semiautomatic and automatic methods. Furthermore, from the validation results it can be obviously noted that the MSD value of our method is much lower than that of other methods. The ability of our method to bring lower MSD could be formed as follows: in our method, there is no risk of edge leakage when region-growing is performed on the binary image obtained through the hybrid image preprocessing. In addition, seed point setting incorporating prior information can make region-growing segment all discrete liver regions, avoiding missing segmentation. Moreover, the unified LSM can resist edge leakage; coupled with the manual termination of the LSM in extreme cases, it is able to receive segmentation results of lower MSD.
Our liver segmentation method requires less interaction than other semiautomatic methods. For instance, in our method, each CT image required 1.6 to 1.8 seed points, but in the method proposed by Yang et al. [
A limitation of the liver segmentation method is that running time has not been optimized, needing to be further reduced. During the segmentation process, the unified LSM propagation spent the most time. So for the future work, we will attempt to apply the narrowband scheme [
Liver tumor segmentation could be regarded as an optimization problem [
The capabilities of the developed liver tumor segmentation method were evaluated with a varied collection of liver tumors. For the local hospital data, most tumors tend to be easily segmented because of obvious boundaries and high contrast. And we used this dataset for method training to determine optimal parameters. The MIDAS dataset provides manual segmentation from five radiologists. Validation with this dataset shows that our method is robust to different manual delineations. Our method was finally evaluated with the 3Dircadb dataset. The received measures were 28.2%, -8.5%, 1.8 mm, 2.4 mm, and 5.8 mm. The obtained RVD was negative, indicating our method tends to obtain undersegmentation for this dataset. Additionally, the comparison with other state-of-the-art methods on the 3Dircadb dataset shows the proposed method is competitive in both accuracy and efficiency.
Interaction required for the proposed tumor segmentation method is definition of the ROI and initialization of the unified LSM. Interaction time consumed was about 26 s per tumor in our study. Besides, the segmentation running time remains modest. It took about 162 s to segment a tumor, most of which was spent by the HMRF-EM algorithm and the unified LSM. The HMRF-EM algorithm is known to be a time-consuming algorithm, especially for handling large-size images [
In summary, we have, respectively, proposed a liver segmentation method and a liver tumor segmentation method. These two methods are mainly grounded on a novel double-information driven unified LSM, which could obtain more accurate segmentation. Validation with various datasets shows our methods provide accurate and reliable approaches for both liver and liver tumor segmentation. Our methods are competitive with previous methods in both accuracy and efficiency and can delineate boundaries that reach a level of accuracy comparable with those of human raters, which allows their adoptions in clinical practice.
The data used to support the findings of this study are available from the corresponding author upon request.
The authors declare that there are no conflicts of interest regarding the publication of this paper.
This work was supported by Natural Science Foundation of Zhejiang Province, China, under Grant no. LY17E050011 and the research project on key technologies of complex surgery for liver resection based on 3D printing that was funded by Ningbo, China, under Grant no. 2015C50025.