Liver Tumor Segmentation Based on Multi-Scale and Self-Attention Mechanism

Liver cancer has the second highest incidence rate among all types of malignant tumors,and currently,its diagnosis heavily depends on doctors’manual labeling of CT scan images,a process that is time-consuming and susceptible to subjective errors.To address the aforementioned issues,we propose an automatic segmentation model for liver and tumors called Res2Swin Unet,which is based on the Unet architecture.The model combines Attention-Res2 and Swin Transformer modules for liver and tumor segmentation,respectively.Attention-Res2 merges multiple feature map parts with an Attention gate via skip connections,while Swin Transformer captures long-range dependencies and models the input globally.And the model uses deep supervision and a hybrid loss function for faster convergence.On the LiTS2017 dataset,it achieves better segmentation performance than other models,with an average Dice coefficient of 97.0%for liver segmentation and 81.2%for tumor segmentation.

Automatic Segmentation of Liver Tumor in CT Images with Deep Convolutional Neural Networks

Liver tumors segmentation from computed tomography (CT) images is an essential task for diagnosis and treatments of liver cancer. However, it is difficult owing to the variability of appearances, fuzzy boundaries, heterogeneous densities, shapes and sizes of lesions. In this paper, an automatic method based on convolutional neural networks (CNNs) is presented to segment lesions from CT images. The CNNs is one of deep learning models with some convolutional filters which can learn hierarchical features from data. We compared the CNNs model to popular machine learning algorithms: AdaBoost, Random Forests (RF), and support vector machine (SVM). These classifiers were trained by handcrafted features containing mean, variance, and contextual features. Experimental evaluation was performed on 30 portal phase enhanced CT images using leave-one-out cross validation. The average Dice Similarity Coefficient (DSC), precision, and recall achieved of 80.06% ± 1.63%, 82.67% ± 1.43%, and 84.34% ± 1.61%, respectively. The results show that the CNNs method has better performance than other methods and is promising in liver tumor segmentation.

Successful robotic radical resection of hepatic echinococcosis located in posterosuperior liver segments

BACKGROUND Radical resection is an important treatment method for hepatic echinococcosis.The posterosuperior segments of the liver remain the most challenging region for laparoscopic or robotic hepatectomy.AIM To demonstrate the safety and preliminary experience of robotic radical resection of cystic and alveolar echinococcosis in posterosuperior liver segments.METHODS A retrospective analysis was conducted on the clinical data of 5 patients with a median age of 37 years(21-56 years)with cystic and alveolar echinococcosis in difficult liver lesions admitted to two centers from September to December 2019.The surgical methods included total pericystectomy,segmental hepatectomy,or hemihepatectomy.RESULTS Among the 5 patients,4 presented with cystic echinococcosis and 1 presented with alveolar echinococcosis,all of whom underwent robotic radical operation successfully without conversion to laparotomy.Total caudate lobectomy was performed in 2 cases,hepatectomy of segment Ⅶ in 1 case,total pericystectomy of segment Ⅷ in 1 case,and right hemihepatectomy in 1 case.Operation time was 225 min(175-300 min);blood loss was 100 mL(50-600 mL);and postoperative hospital stay duration was 10 d(5-19 d).The Clavien-Dindo complication grade was Ⅰ in 4 cases and Ⅱ in 1 case.No recurrence of echinococcosis was found in any patient at the 3 mo of follow-up.CONCLUSION Robotic radical surgery for cystic and selected alveolar echinococcosis in posterosuperior liver segments is safe and feasible.

Review on the Methods of Automatic Liver Segmentation from Abdominal Images

Automatic liver segmentation from abdominal images is challenging on the aspects of segmentation accuracy, automation and robustness. There exist many methods of liver segmentation and ways of categorisingthem. In this paper, we present a new way of summarizing the latest achievements in automatic liver segmentation. We categorise a segmentation method according to the image feature it works on, therefore better summarising the performance of each category and leading to finding an optimal solution for a particular segmentation task. All the methods of liver segmentation are categorized into three main classes including gray level based method, structure based method and texture based method. In each class, the latest advance is reviewed with summary comments on the advantages and drawbacks of each discussed approach. Performance comparisons among the classes are given along with the remarks on the problems existed and possible solutions. In conclusion, we point out that liver segmentation is still an open issue and the tendency is that multiple methods will be employed together to achieve better segmentation performance.

Automatic liver and tumor segmentation based on deep learning and globally optimized refinement

Automatic segmentation of the liver and hepatic lesions from abdominal 3D comput-ed tomography(CT)images is fundamental tasks in computer-assisted liver surgery planning.However,due to complex backgrounds,ambiguous boundaries,heterogeneous appearances and highly varied shapes of the liver,accurate liver segmentation and tumor detection are stil-1 challenging problems.To address these difficulties,we propose an automatic segmentation framework based on 3D U-net with dense connections and globally optimized refinement.First-ly,a deep U-net architecture with dense connections is trained to learn the probability map of the liver.Then the probability map goes into the following refinement step as the initial surface and prior shape.The segmentation of liver tumor is based on the similar network architecture with the help of segmentation results of liver.In order to reduce the infuence of the surrounding tissues with the similar intensity and texture behavior with the tumor region,during the training procedure,I x liverlabel is the input of the network for the segmentation of liver tumor.By do-ing this,the accuracy of segmentation can be improved.The proposed method is fully automatic without any user interaction.Both qualitative and quantitative results reveal that the pro-posed approach is efficient and accurate for liver volume estimation in clinical application.The high correlation between the automatic and manual references shows that the proposed method can be good enough to replace the time-consuming and non-reproducible manual segmentation method.

Improvement of Liver Segmentation by Combining High Order Statistical Texture Features with Anatomical Structural Features

Automatic segmentation of liver in medical images is challenging on the aspects of accuracy, automation and robustness. A crucial stage of the liver segmentation is the selection of the image features for the segmentation. This paper presents an accurate liver segmentation algorithm. The approach starts with a texture analysis which results in an optimal set of texture features including high order statistical texture features and anatomical structural features. Then, it creates liver distribution image by classifying the original image pixelwisely using support vector machines. Lastly, it uses a group of morphological operations to locate the liver organ accurately in the image. The novelty of the approach is resided in the fact that the features are so selected that both local and global texture distributions are considered, which is important in liver organ segmentation where neighbouring tissues and organs have similar greyscale distributions. Experiment results of liver segmentation on CT images using the proposed method are presented with performance validation and discussion.

Empirical Comparisons of Deep Learning Networks on Liver Segmentation

Accurate segmentation of CT images of liver tumors is an important adjunct for the liver diagnosis and treatment of liver diseases.In recent years,due to the great improvement of hard device,many deep learning based methods have been proposed for automatic liver segmentation.Among them,there are the plain neural network headed by FCN and the residual neural network headed by Resnet,both of which have many variations.They have achieved certain achievements in medical image segmentation.In this paper,we firstly select five representative structures,i.e.,FCN,U-Net,Segnet,Resnet and Densenet,to investigate their performance on liver segmentation.Since original Resnet and Densenet could not perform image segmentation directly,we make some adjustments for them to perform live segmentation.Our experimental results show that Densenet performs the best on liver segmentation,followed by Resnet.Both perform much better than Segnet,U-Net,and FCN.Among Segnet,U-Net,and FCN,U-Net performs the best,followed by Segnet.FCN performs the worst.

An Efficient Liver-Segmentation System Based on a Level-Set Method and Consequent Processes

This paper presents an efficient liver-segmentation system developed by combining three ideas under the operations of a level-set method and consequent processes. First, an effective initial process creates mask and seed regions. The mask regions assist in prevention of leakage regions due to an overlap of gray-intensities between liver and another soft-tissue around ribs and verte-brae. The seed regions are allocated inside the liver to measure statistical values of its gray-intensities. Second, we introduce liver-corrective images to represent statistical regions of the liver and preserve edge information. These images help a geodesic active contour (GAC) to move without obstruction from high level of image noises. Lastly, the computation time in a level-set based on reaction-diffusion evolution and the GAC method is reduced by using a concept of multi-resolution. We applied the proposed system to 40 sets of 3D CT-liver data, which were acquired from four patients (10 different sets per patient) by a 4D-CT imaging system. The segmentation results showed 86.38% ± 4.26% (DSC: 91.38% ± 2.99%) of similarities to outlines of manual delineation provided by a radiologist. Meanwhile, the results of liver segmentation only using edge images presented 79.17% ± 5.15% or statistical regions showed 74.04% ± 9.77% of similarities.

Liver resection vs radiofrequency ablation in single hepatocellular carcinoma of posterosuperior segments in elderly patients

BACKGROUND Liver resection and radiofrequency ablation are considered curative options for hepatocellular carcinoma.The choice between these techniques is still controversial especially in cases of hepatocellular carcinoma affecting posterosuperior segments in elderly patients.AIM To compare post-operative outcomes between liver resection and radiofrequency ablation in elderly with single hepatocellular carcinoma located in posterosuperior segments.METHODS A retrospective multicentric study was performed enrolling 77 patients age≥70-years-old with single hepatocellular carcinoma(≤30 mm),located in posterosuperior segments(4a,7,8).Patients were divided into liver resection and radiofrequency ablation groups and preoperative,peri-operative and long-term outcomes were retrospectively analyzed and compared using a 1:1 propensity score matching.RESULTS After propensity score matching,twenty-six patients were included in each group.Operative time and overall postoperative complications were higher in the resection group compared to the ablation group(165 min vs 20 min,P<0.01;54%vs 19%P=0.02 respectively).A median hospital stay was significantly longer in the resection group than in the ablation group(7.5 d vs 3 d,P<0.01).Ninety-day mortality was comparable between the two groups.There were no significant differences between resection and ablation group in terms of overall survival and disease free survival at 1,3,and 5 years.CONCLUSION Radiofrequency ablation in posterosuperior segments in elderly is safe and feasible and ensures a short hospital stay,better quality of life and does not modify the overall and disease-free survival.

Liver Segmentation from CT Image Using Fuzzy Clustering and Level Set

This paper presents a fully automatic segmentation method of liver CT scans using fuzzy c-mean clustering and level set. First, the contrast of original image is enhanced to make boundaries clearer;second, a spatial fuzzy c-mean clustering combining with anatomical prior knowledge is employed to extract liver region automatically;thirdly, a distance regularized level set is used for refinement;finally, morphological operations are used as post-processing. The experiment result shows that the method can achieve high accuracy (0.9986) and specificity (0.9989). Comparing with standard level set method, our method is more effective in dealing with over-segmentation problem.

Single hepatocellular carcinoma ≤ 3 cm in left lateral segment:Liver resection or radiofrequency ablation?

AIM:To evaluate the long-term results of radiofrequency ablation(RFA)compared to left lateral sectionectomy(LLS)in patients with Child-Pugh class A disease for the treatment of single and small hepatocellular carcinoma(HCC)in the left lateral segments.METHODS:We retrospectively reviewed the data of133 patients with single HCC(≤3 cm)in their left lateral segments who underwent curative LLS(n=66)or RFA(n=67)between 2006 and 2010.RESULTS:The median follow-up period was 33.5mo in the LLS group and 29 mo in the RFA group(P=0.060).Most patients had hepatitis B virus-related HCC.The hospital stay was longer in the LLS group than in the RFA group(8 d vs 2 d,P<0.001).The 1-,2-,and 3-year disease-free survival and overall survival rates were 80.0%,68.2%,and 60.0%,and 95.4%,92.3%,and 92.3%,respectively,for the LLS group;and 80.8%,59.9%,and 39.6%,and 98.2%,92.0%,and 74.4%,respectively,for the RFA group.The disease-free survival curve and overall survival curve were higher in the LLS group than in the RFA group(P=0.012 and P=0.013,respectively).Increased PIVKA-Ⅱlevels and small tumor size were associated with HCC recurrence in multivariate analysis.CONCLUSION:Liver resection is suitable for single HCC≤3 cm in the left lateral segments.

Automatic Liver Segmentation Scheme for MRI Images Based on Cellular Neural Networks

Currently,the processing speed of existing automatic liver segmentation for Magnetic Resonance Imaging (MRI) images is relatively slow.An automatic liver segmentation scheme for MRI images based on Cellular Neural Networks (CNN) is presented in this paper.It ensures the validity of this scheme and at the same time completes the image segmentation faster to accurately calculate the liver volume by using parallel computing in real time.In order to facilitate the CNN image processing,firstly,three-dimensional liver MRI images should be transformed into binary images;secondly,an appropriate template parameter of the Global Connectivity Detection CNN (GCD CNN) shall be selected to probe the connectivity of the liver to extract the entire liver;and then the Hole-Filler CNN (HF CNN) are used to repair the entire extracting liver and improve the accuracy of liver segmentation;finally,the liver volume is obtained.Results show that the scheme can ensure the accuracy of the automatic segmentation of the liver,and it can also improve the processing speed at the same time.The liver volume calculated is in line with the clinical diagnosis.

Stepwise evaluation of liver sectors and liver segments by endoscopic ultrasound

The liver has eight segments, which are referred to by numbers or by names. The numbering of the segments is done in a counterclockwise manner with the liver being viewed from the inferior surface, starting from Segment Ⅰ(the caudate lobe). Standard anatomical description of the liver segments is available by computed tomographic scan and ultrasonography. Endoscopic ultrasound(EUS) has been used for a detailed imaging of many intra-abdominal organs and for the assessment of intra-abdominal vasculature. A stepwise evaluation of the liver segments by EUS has not been described. In this article, we have described a stepwise evaluation of the liver segments by EUS. This information can be useful for planning successful radical surgeries, preparing for biopsy, portal vein embolization, transjugular intrahepatic portosystemic shunt, tumour resection or partial hepatectomy, and for planning EUS guided diagnostic and therapeutic procedures.

Automatic Segmentation of Liver from Abdominal Computed Tomography Images Using Energy Feature

Liver Segmentation is one of the challenging tasks in detecting and classifying liver tumors from Computed Tomography(CT)images.The segmentation of hepatic organ is more intricate task,owing to the fact that it possesses a sizeable quantum of vascularization.This paper proposes an algorithm for automatic seed point selection using energy feature for use in level set algorithm for segmentation of liver region in CT scans.The effectiveness of the method can be determined when used in a model to classify the liver CT images as tumorous or not.This involves segmentation of the region of interest(ROI)from the segmented liver,extraction of the shape and texture features from the segmented ROI and classification of the ROIs as tumorous or not by using a classifier based on the extracted features.In this work,the proposed seed point selection technique has been used in level set algorithm for segmentation of liver region in CT scans and the ROIs have been extracted using Fuzzy C Means clustering(FCM)which is one of the algorithms to segment the images.The dataset used in this method has been collected from various repositories and scan centers.The outcome of this proposed segmentation model has reduced the area overlap error that could offer the intended accuracy and consistency.It gives better results when compared with other existing algorithms.Fast execution in short span of time is another advantage of this method which in turns helps the radiologist to ascertain the abnormalities instantly.

Topological approach of liver segmentation based on 3D visualization technology in surgical planning for split liver transplantation

BACKGROUND Split liver transplantation(SLT)is a complex procedure.The left-lateral and right tri-segment splits are the most common surgical approaches and are based on the Couinaud liver segmentation theory.Notably,the liver surface following right trisegment splits may exhibit different degrees of ischemic changes related to the destruction of the local portal vein blood flow topology.There is currently no consensus on preoperative evaluation and predictive strategy for hepatic segmental necrosis after SLT.AIM To investigate the application of the topological approach in liver segmentation based on 3D visualization technology in the surgical planning of SLT.METHODS Clinical data of 10 recipients and 5 donors who underwent SLT at Shenzhen Third People’s Hospital from January 2020 to January 2021 were retrospectively analyzed.Before surgery,all the donors were subjected to 3D modeling and evaluation.Based on the 3D-reconstructed models,the liver splitting procedure was simulated using the liver segmentation system described by Couinaud and a blood flow topology liver segmentation(BFTLS)method.In addition,the volume of the liver was also quantified.Statistical indexes mainly included the hepatic vasculature and expected volume of split grafts evaluated by 3D models,the actual liver volume,and the ischemia state of the hepatic segments during the actual surgery.RESULTS Among the 5 cases of split liver surgery,the liver was split into a left-lateral segment and right trisegment in 4 cases,while 1 case was split using the left and right half liver splitting.All operations were successfully implemented according to the preoperative plan.According to Couinaud liver segmentation system and BFTLS methods,the volume of the left lateral segment was 359.00±101.57 mL and 367.75±99.73 mL,respectively,while that measured during the actual surgery was 397.50±37.97 mL.The volume of segment IV(the portion of ischemic liver lobes)allocated to the right tri-segment was 136.31±86.10 mL,as determined using the topological approach to liver segmentation.However,during the actual surgical intervention,ischemia of the right tri-segment section was observed in 4 cases,including 1 case of necrosis and bile leakage,with an ischemic liver volume of 238.7 mL.CONCLUSION 3D visualization technology can guide the preoperative planning of SLT and improve accuracy during the intervention.The simulated operation based on 3D visualization of blood flow topology may be useful to predict the degree of ischemia in the liver segment and provide a reference for determining whether the ischemic liver tissue should be removed during the surgery.

Hepatic vessel segmentation on contrast enhanced CT image sequence for liver transplantation planning

The structure and morphology of the hepatic vessels and their relationship between tumors and liver segments are major interests to surgeons for liver surgical planning. In case of living donor liver transplantation (LDLT), the most important step in determining donor suitability is an accurate assessment of the liver volume available for transplantation. In addition, the mutual principles of the procedures include dissection in the appropriate anatomic plane without portal occlusion, minimization of blood loss, and avoidance of injury to the remaining liver. It is essential first step to identify and evaluate the major hepatic vascular structure for liver surgical planning. In this paper, the threshold was determined to segment the liver region automatically based on the distribution ratio of intensity value;and the hepatic vessels were extracted with mathematical morphology transformation, which called hit operation, that is slightly modified version of hit-and-miss operation on contrast enhanced CT image sequence. We identified the vein using the preserved voxel connectivity between two consecutive transverse image sequences, followed by resection into right lobe including right hepatic vein, middle hepatic vein branches andleft lobe including left hepatic vein. An automated hepatic vessel segmentation scheme is recommended for liver surgical planning such as tumor resection and transplantation. These vessel extraction method combined with liver region segmentation technique could be applicable to extract tree-like organ structures such as carotid, renal, coronary artery, and airway path from various medical imaging modalities.

Liver Segmentation in CT Images Based on DRLSE Model

Liver segmentation in CT images is an important step for liver volumetry and vascular evaluation in liver pre-surgical planning. In this paper, a segmentation method based on distance regularized level set evolution(DRLSE) model was proposed, which incorporated a distance regularization term into the conventional Chan-Vese (C-V) model. In addition, the region growing method was utilized to generate the initial liver mask for each slice, which could decrease the computation time for level-set propagation. The experimental results show that the method can dramatically decrease the evolving time and keep the accuracy of segmentation. The new method is averagely 15 times faster than the method based on conventional C-V model in segmenting a slice.

Automatic Liver Tumor Segmentation in CT Modalities Using MAT-ACM

In the recent days, the segmentation of Liver Tumor (LT) has beendemanding and challenging. The process of segmenting the liver and accuratelyspotting the tumor is demanding due to the diversity of shape, texture, and intensity of the liver image. The intensity similarities of the neighboring organs of theliver create difficulties during liver segmentation. The manual segmentation doesnot provide an accurate segmentation because the results provided by differentmedical experts can vary. Also, this manual technique requires a large numberof image slices and time for segmentation. To solve these issues, the Fully Automatic Segmentation (FAS) technique is proposed. In this proposed Multi-AngleTexture Active Contour Model (MAT-ACM) method, the input Computed Tomography (CT) image is preprocessed by Contrast Enhancement (CE) with Non-Linear Mapping Technique (NLMT), in which the liver is differentiated from itsneighbouring soft tissues with related strength. Then, the filtered images are givenas the input to Adaptive Edge Modeling (AEM) with Canny Edge Detection(CED) technique, which segments the Liver Region (LR) from the given CTimages. An AEM with a CED model is implemented, which increases the convergence speed of the iterative process for decreasing the Volumetric Overlap Error(VOE) is 6.92% rates when compared with the traditional Segmentation Techniques (ST). Finally, the Liver Tumor Segmentation (LTS) is developed by applyingthe MAT-ACM, which accurately segments the LR from the segmented LRs. Theevaluation of the proposed method is compared with the existing LTS methodsusing various performance measures to prove the superiority of the proposedMAT-ACM method.

Chinese expert consensus on laparoscopic hepatic segmentectomy and subsegmentectomy navigated by augmented-and mixed-reality technology combined with indocyanine green fluorescence imaging

Augmented-and mixed-reality technologies have pioneered the realization of real-time fusion and interactive projection for laparoscopic surgeries.Indocyanine green fluorescence imaging technology has enabled anatomical,functional,and radical hepatectomy through tumor identification and localization of target hepatic segments,driving a transformative shift in themanagement of hepatic surgical diseases,moving away from traditional,empirical diagnostic and treatment approaches toward digital,intelligent ones.The Hepatic Surgery Group of the Surgery Branch of the Chinese Medical Association,Digital Medicine Branch of the Chinese Medical Association,Digital Intelligent Surgery Committee of the Chinese Society of ResearchHospitals,and Liver Cancer Committee of the Chinese Medical Doctor Association organized the relevant experts in China to formulate this consensus.This consensus provides a comprehensive outline of the principles,advantages,processes,and key considerations associated with the application of augmented reality and mixed-reality technology combined with indocyanine green fluorescence imaging technology for hepatic segmental and subsegmental resection.The purpose is to streamline and standardize the application of these technologies.

Liver Hydatid CT Image Segmentation Using Smoothed Bayesian Classification Method and Modified Parametric Active Contour Model

Liver hydatid disease is a common parasitic disease in farm and pastoral areas, which seriously influences people's health. Based on CT imaging features of this disease, an iterative approach for liver segmentation and hydatid lesion extraction simultaneously is proposed. In each iteration, our algorithm consists of two main steps: 1) according to the user-defined pixel seeds in the liver and hydatid lesion, Gaussian probability model fitting and smoothed Bayesian classification are applied to get initial segmentation of liver and lesion; 2) the parametric active contour model using priori shape force field is adopted to refine initial segmentation. We make subjective and objective evaluation on the proposed algorithm validity by the experiments of liver and hydatid lesion segmentation on different patients' CT slices. In comparison with ground-truth manual segmentation results, the experimental results show the effectiveness of our method to segment liver and hydatid lesion.