Neuronal soma segmentation plays a crucial role in neuroscience applications.However,the fine structure,such as boundaries,small-volume neuronal somata and fibers,are commonly present in cell images,which pose a chall...Neuronal soma segmentation plays a crucial role in neuroscience applications.However,the fine structure,such as boundaries,small-volume neuronal somata and fibers,are commonly present in cell images,which pose a challenge for accurate segmentation.In this paper,we propose a 3D semantic segmentation network for neuronal soma segmentation to address this issue.Using an encoding-decoding structure,we introduce a Multi-Scale feature extraction and Adaptive Weighting fusion module(MSAW)after each encoding block.The MSAW module can not only emphasize the fine structures via an upsampling strategy,but also provide pixel-wise weights to measure the importance of the multi-scale features.Additionally,a dynamic convolution instead of normal convolution is employed to better adapt the network to input data with different distributions.The proposed MSAW-based semantic segmentation network(MSAW-Net)was evaluated on three neuronal soma images from mouse brain and one neuronal soma image from macaque brain,demonstrating the efficiency of the proposed method.It achieved an F1 score of 91.8%on Fezf2-2A-CreER dataset,97.1%on LSL-H2B-GFP dataset,82.8%on Thy1-EGFP-Mline dataset,and 86.9%on macaque dataset,achieving improvements over the 3D U-Net model by 3.1%,3.3%,3.9%,and 2.3%,respectively.展开更多
Image technology is applied more and more to help doctors to improve the accuracy of tumor diagnosis as well as researchers to study tumor characteristics. Image segmentation technology is an important part of image t...Image technology is applied more and more to help doctors to improve the accuracy of tumor diagnosis as well as researchers to study tumor characteristics. Image segmentation technology is an important part of image treatment. This paper summarizes the advances of image segmentation by using artificial neural network including mainly the BP network and convolutional neural network (CNN). Many CNN models with different structures have been built and successfully used in segmentation of tumor images such as supervised and unsupervised learning CNN. It is shown that the application of artificial network can improve the efficiency and accuracy of segmentation of tumor image. However, some deficiencies of image segmentation by using artificial neural network still exist. For example, new methods should be found to reduce the cost of building the marked data set. New artificial networks with higher efficiency should be built.展开更多
Based on a model of network encoding and dynamics called the artificial genome, we propose a segmental duplication and divergence model for evolving artificial regulatory networks. We find that this class of networks ...Based on a model of network encoding and dynamics called the artificial genome, we propose a segmental duplication and divergence model for evolving artificial regulatory networks. We find that this class of networks share structural properties with natural transcriptional regulatory networks. Specifically, these networks can display scale-free and small-world structures. We also find that these networks have a higher probability to operate in the ordered regimen, and a lower probability to operate in the chaotic regimen. That is, the dynamics of these networks is similar to that of natural networks. The results show that the structure and dynamics inherent in natural networks may be in part due to their method of generation rather than being exclusively shaped by subsequent evolution under natural selection.展开更多
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, het...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.展开更多
This paper proposes a hybrid technique for color image segmentation. First an input image is converted to the image of CIE L＊a＊b＊ color space. The color features ＂a＂ and ＂b＂ of CIE L^＊a^＊b^＊ are then fed int...This paper proposes a hybrid technique for color image segmentation. First an input image is converted to the image of CIE L＊a＊b＊ color space. The color features ＂a＂ and ＂b＂ of CIE L^＊a^＊b^＊ are then fed into fuzzy C-means （FCM） clustering which is an unsupervised method. The labels obtained from the clustering method FCM are used as a target of the supervised feed forward neural network. The network is trained by the Levenberg-Marquardt back-propagation algorithm, and evaluates its performance using mean square error and regression analysis. The main issues of clustering methods are determining the number of clusters and cluster validity measures. This paper presents a method namely co-occurrence matrix based algorithm for finding the number of clusters and silhouette index values that are used for cluster validation. The proposed method is tested on various color images obtained from the Berkeley database. The segmentation results from the proposed method are validated and the classification accuracy is evaluated by the parameters sensitivity, specificity, and accuracy.展开更多
This paper concerns the problem of object segmentation in real-time for picking system. A region proposal method inspired by human glance based on the convolutional neural network is proposed to select promising regio...This paper concerns the problem of object segmentation in real-time for picking system. A region proposal method inspired by human glance based on the convolutional neural network is proposed to select promising regions, allowing more processing is reserved only for these regions. The speed of object segmentation is significantly improved by the region proposal method.By the combination of the region proposal method based on the convolutional neural network and superpixel method, the category and location information can be used to segment objects and image redundancy is significantly reduced. The processing time is reduced considerably by this to achieve the real time. Experiments show that the proposed method can segment the interested target object in real time on an ordinary laptop.展开更多
Moving object segmentation (MOS) is one of the essential functions of the vision system of all robots,including medical robots. Deep learning-based MOS methods, especially deep end-to-end MOS methods, are actively inv...Moving object segmentation (MOS) is one of the essential functions of the vision system of all robots,including medical robots. Deep learning-based MOS methods, especially deep end-to-end MOS methods, are actively investigated in this field. Foreground segmentation networks (FgSegNets) are representative deep end-to-endMOS methods proposed recently. This study explores a new mechanism to improve the spatial feature learningcapability of FgSegNets with relatively few brought parameters. Specifically, we propose an enhanced attention(EA) module, a parallel connection of an attention module and a lightweight enhancement module, with sequentialattention and residual attention as special cases. We also propose integrating EA with FgSegNet_v2 by taking thelightweight convolutional block attention module as the attention module and plugging EA module after the twoMaxpooling layers of the encoder. The derived new model is named FgSegNet_v2 EA. The ablation study verifiesthe effectiveness of the proposed EA module and integration strategy. The results on the CDnet2014 dataset,which depicts human activities and vehicles captured in different scenes, show that FgSegNet_v2 EA outperformsFgSegNet_v2 by 0.08% and 14.5% under the settings of scene dependent evaluation and scene independent evaluation, respectively, which indicates the positive effect of EA on improving spatial feature learning capability ofFgSegNet_v2.展开更多
Clustering algorithms in feature space are important methods in image segmentation. The choice of the effective feature parameters and the construction of the clustering method are key problems encountered with cluste...Clustering algorithms in feature space are important methods in image segmentation. The choice of the effective feature parameters and the construction of the clustering method are key problems encountered with clustering algorithms. In this paper, the multifractal dimensions are chosen as the segmentation feature parameters which are extracted from original image and wavelet-transformed image. SOM (Self-Organizing Map) network is applied to cluster the segmentation feature parameters. The experiment shows that the performance of the presented algorithm is very good.展开更多
In this paper, we propose several improved neural networks and training strategy using data augmentation to segment human radius accurately and efficiently. This method can provide pixel-level segmentation accuracy th...In this paper, we propose several improved neural networks and training strategy using data augmentation to segment human radius accurately and efficiently. This method can provide pixel-level segmentation accuracy through the low-level features of the neural network, and automatically distinguish the classification of radius. The versatility and applicability can be effectively improved by learning and training digital X-ray images obtained from digital X-ray imaging systems of different manufacturers.展开更多
neural network model based on backbone propagation was applied to Learn-ing and predicting the interaction between antiparallelly interactive peptide seg-ments in proteins.Hydrophobic properties pf residues were found...neural network model based on backbone propagation was applied to Learn-ing and predicting the interaction between antiparallelly interactive peptide seg-ments in proteins.Hydrophobic properties pf residues were found dominant in in-terpeptides.Weights of each kind of residues, obtained by this work,suggestedsome different scales for the hydrophobicity of the residue.These will be helpful in understanding polypeptide structure and protein folding.展开更多
AIM: To explore a segmentation algorithm based on deep learning to achieve accurate diagnosis and treatment of patients with retinal fluid.METHODS: A two-dimensional(2D) fully convolutional network for retinal segment...AIM: To explore a segmentation algorithm based on deep learning to achieve accurate diagnosis and treatment of patients with retinal fluid.METHODS: A two-dimensional(2D) fully convolutional network for retinal segmentation was employed. In order to solve the category imbalance in retinal optical coherence tomography(OCT) images, the network parameters and loss function based on the 2D fully convolutional network were modified. For this network, the correlations of corresponding positions among adjacent images in space are ignored. Thus, we proposed a three-dimensional(3D) fully convolutional network for segmentation in the retinal OCT images.RESULTS: The algorithm was evaluated according to segmentation accuracy, Kappa coefficient, and F1 score. For the 3D fully convolutional network proposed in this paper, the overall segmentation accuracy rate is 99.56%, Kappa coefficient is 98.47%, and F1 score of retinal fluid is 95.50%. CONCLUSION: The OCT image segmentation algorithm based on deep learning is primarily founded on the 2D convolutional network. The 3D network architecture proposed in this paper reduces the influence of category imbalance, realizes end-to-end segmentation of volume images, and achieves optimal segmentation results. The segmentation maps are practically the same as the manual annotations of doctors, and can provide doctors with more accurate diagnostic data.展开更多
The multilayer feedforward network is used for image segmentation. This paper deals with the procedure of achieving the learning patterns and the method of improving the learning rate. The experiment shows that the im...The multilayer feedforward network is used for image segmentation. This paper deals with the procedure of achieving the learning patterns and the method of improving the learning rate. The experiment shows that the image segmentation can get better result from using the multilayer feedforward network.展开更多
This paper presents an unsupervised range image segmentation based on Kohonen neural network. At first, the derivative and partial derivative of each point are calculated and the normal in each points is gotten. With ...This paper presents an unsupervised range image segmentation based on Kohonen neural network. At first, the derivative and partial derivative of each point are calculated and the normal in each points is gotten. With the character vectors including normal and range value, self-organization map is introduced to cluster. The normal analysis is used to eliminate over-segmentation and the last result is gotten. This method avoid selecting original seeds and uses fewer samples, moreover computes rapidly. The experiment shows the better performance.展开更多
A new approach for multilevel image segmentation based on fuzzy cellular neural network(CNN) is proposed. Based on a novel fuzzy CNN, a new template is proposed for multilevel image segmentation. The result of compute...A new approach for multilevel image segmentation based on fuzzy cellular neural network(CNN) is proposed. Based on a novel fuzzy CNN, a new template is proposed for multilevel image segmentation. The result of computer simulation proves this approach is reasonable. The stability of the fuzzy neural network is also analyzed in this paper.展开更多
This paper presents a new solution to the image segmentation problem, which is based on fuzzy-neural-network hybrid system (FNNHS). This approach can use the experiential knowledge and the ability of neural networks w...This paper presents a new solution to the image segmentation problem, which is based on fuzzy-neural-network hybrid system (FNNHS). This approach can use the experiential knowledge and the ability of neural networks which learn knowledge from the examples, to obtain the well performed fuzzy rules. Furthermore this fuzzy inference system is completed by neural network structure which can work in parallel. The segmentation process consists of pre-segmentation based on region growing algorithm and region merging based on FNNHS. The experimental results on the complicated image manifest the utility of this method.展开更多
BACKGROUND Automated,accurate,objective,and quantitative medical image segmentation has remained a challenging goal in computer science since its inception.This study applies the technique of convolutional neural netw...BACKGROUND Automated,accurate,objective,and quantitative medical image segmentation has remained a challenging goal in computer science since its inception.This study applies the technique of convolutional neural networks(CNNs)to the task of segmenting carotid arteries to aid in the assessment of pathology.AIM To investigate CNN’s utility as an ancillary tool for researchers who require accurate segmentation of carotid vessels.METHODS An expert reader delineated vessel wall boundaries on 4422 axial T2-weighted magnetic resonance images of bilateral carotid arteries from 189 subjects with clinically evident atherosclerotic disease.A portion of this dataset was used to train two CNNs(one to segment the vessel lumen and the other to segment the vessel wall)with the remaining portion used to test the algorithm’s efficacy by comparing CNN segmented images with those of an expert reader.Overall quantitative assessment between automated and manual segmentations was determined by computing the DICE coefficient for each pair of segmented images in the test dataset for each CNN applied.The average DICE coefficient for the test dataset(CNN segmentations compared to expert’s segmentations)was 0.96 for the lumen and 0.87 for the vessel wall.Pearson correlation values and the intra-class correlation coefficient(ICC)were computed for the lumen(Pearson=0.98,ICC=0.98)and vessel wall(Pearson=0.88,ICC=0.86)segmentations.Bland-Altman plots of area measurements for the CNN and expert readers indicate good agreement with a mean bias of 1%-8%.CONCLUSION Although the technique produces reasonable results that are on par with expert human assessments,our application requires human supervision and monitoring to ensure consistent results.We intend to deploy this algorithm as part of a software platform to lessen researchers’workload to more quickly obtain reliable results.展开更多
An image segmentation algorithm of the restrained fuzzy Kohonen clustering network （RFKCN） based on high- dimension fuzzy character is proposed. The algorithm includes two steps. The first step is the fuzzification ...An image segmentation algorithm of the restrained fuzzy Kohonen clustering network （RFKCN） based on high- dimension fuzzy character is proposed. The algorithm includes two steps. The first step is the fuzzification of pixels in which two redundant images are built by fuzzy mean value and fuzzy median value. The second step is to construct a three-dimensional （3-D） feature vector of redundant images and their original images and cluster the feature vector through RFKCN, to realize image seg- mentation. The proposed algorithm fully takes into account not only gray distribution information of pixels, but also relevant information and fuzzy information among neighboring pixels in constructing 3- D character space. Based on the combination of competitiveness, redundancy and complementary of the information, the proposed algorithm improves the accuracy of clustering. Theoretical anal- yses and experimental results demonstrate that the proposed algorithm has a good segmentation performance.展开更多
<strong>Purpose</strong><span style="font-family:;" "=""><span style="font-family:Verdana;"><strong>: </strong></span><span style=&q...<strong>Purpose</strong><span style="font-family:;" "=""><span style="font-family:Verdana;"><strong>: </strong></span><span style="font-family:Verdana;">To improve the liver auto-segmentation performance of three-</span><span style="font-family:Verdana;">dimensional (3D) U-net by replacing the conventional up-sampling convolution layers with the Pixel De-convolutional Network (PDN) that considers spatial features. </span><b><span style="font-family:Verdana;">Methods</span></b><span style="font-family:Verdana;">: The U-net was originally developed to segment neuronal structure with outstanding performance but suffered serious artifacts from indirectly unrelated adjacent pixels in its up-sampling layers. The hypothesis of this study was that the segmentation quality of </span></span><span style="font-family:Verdana;">the </span><span style="font-family:Verdana;">liver could be improved with PDN in which the up-sampling layer was replaced by a pixel de-convolution layer (PDL). Seventy</span><span style="font-family:Verdana;">-</span><span style="font-family:;" "=""><span style="font-family:Verdana;">eight plans of abdominal cancer patients were anonymized and exported. Sixty-two were chosen for training two networks: 1) 3D U-Net, and 2) 3D PDN, by minimizing the Dice loss function. The other sixteen plans were used to test the performance. The similarity Dice and Average Hausdorff Distance (AHD) were calculated and compared between these two networks. </span><b><span style="font-family:Verdana;">Results</span></b><span style="font-family:Verdana;">: The computation time for 62 training cases and 200 training epochs was about 30 minutes for both networks. The segmentation performance was evaluated using the remaining 16 cases. For the Dice score, the mean ± standard deviation were 0.857 ± 0.011 and 0.858 ± 0.015 for the PDN and U-Net, respectively. For the AHD, the mean ± standard deviation were 1.575 ± 0.373 and 1.675 ± 0.769, respectively, corresponding to an improvement of 6.0% and 51.5% of mean and standard deviation for the PDN. </span><b><span style="font-family:Verdana;">Conclusion</span></b><span style="font-family:Verdana;">: The PDN has outperformed the U-Net on liver auto-segmentation. The predicted contours of PDN are more conformal and smoother when compared with</span></span><span style="font-family:Verdana;"> the</span><span style="font-family:Verdana;"> U-Net.</span>展开更多
基金supported by the STI2030-Major-Projects(No.2021ZD0200104)the National Natural Science Foundations of China under Grant 61771437.
文摘Neuronal soma segmentation plays a crucial role in neuroscience applications.However,the fine structure,such as boundaries,small-volume neuronal somata and fibers,are commonly present in cell images,which pose a challenge for accurate segmentation.In this paper,we propose a 3D semantic segmentation network for neuronal soma segmentation to address this issue.Using an encoding-decoding structure,we introduce a Multi-Scale feature extraction and Adaptive Weighting fusion module(MSAW)after each encoding block.The MSAW module can not only emphasize the fine structures via an upsampling strategy,but also provide pixel-wise weights to measure the importance of the multi-scale features.Additionally,a dynamic convolution instead of normal convolution is employed to better adapt the network to input data with different distributions.The proposed MSAW-based semantic segmentation network(MSAW-Net)was evaluated on three neuronal soma images from mouse brain and one neuronal soma image from macaque brain,demonstrating the efficiency of the proposed method.It achieved an F1 score of 91.8%on Fezf2-2A-CreER dataset,97.1%on LSL-H2B-GFP dataset,82.8%on Thy1-EGFP-Mline dataset,and 86.9%on macaque dataset,achieving improvements over the 3D U-Net model by 3.1%,3.3%,3.9%,and 2.3%,respectively.
文摘Image technology is applied more and more to help doctors to improve the accuracy of tumor diagnosis as well as researchers to study tumor characteristics. Image segmentation technology is an important part of image treatment. This paper summarizes the advances of image segmentation by using artificial neural network including mainly the BP network and convolutional neural network (CNN). Many CNN models with different structures have been built and successfully used in segmentation of tumor images such as supervised and unsupervised learning CNN. It is shown that the application of artificial network can improve the efficiency and accuracy of segmentation of tumor image. However, some deficiencies of image segmentation by using artificial neural network still exist. For example, new methods should be found to reduce the cost of building the marked data set. New artificial networks with higher efficiency should be built.
文摘Based on a model of network encoding and dynamics called the artificial genome, we propose a segmental duplication and divergence model for evolving artificial regulatory networks. We find that this class of networks share structural properties with natural transcriptional regulatory networks. Specifically, these networks can display scale-free and small-world structures. We also find that these networks have a higher probability to operate in the ordered regimen, and a lower probability to operate in the chaotic regimen. That is, the dynamics of these networks is similar to that of natural networks. The results show that the structure and dynamics inherent in natural networks may be in part due to their method of generation rather than being exclusively shaped by subsequent evolution under natural selection.
文摘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.
文摘This paper proposes a hybrid technique for color image segmentation. First an input image is converted to the image of CIE L＊a＊b＊ color space. The color features ＂a＂ and ＂b＂ of CIE L^＊a^＊b^＊ are then fed into fuzzy C-means （FCM） clustering which is an unsupervised method. The labels obtained from the clustering method FCM are used as a target of the supervised feed forward neural network. The network is trained by the Levenberg-Marquardt back-propagation algorithm, and evaluates its performance using mean square error and regression analysis. The main issues of clustering methods are determining the number of clusters and cluster validity measures. This paper presents a method namely co-occurrence matrix based algorithm for finding the number of clusters and silhouette index values that are used for cluster validation. The proposed method is tested on various color images obtained from the Berkeley database. The segmentation results from the proposed method are validated and the classification accuracy is evaluated by the parameters sensitivity, specificity, and accuracy.
基金supported by the National Natural Science Foundation of China(61233010 61305106)+2 种基金the Shanghai Natural Science Foundation(17ZR1409700 18ZR1415300)the basic research project of Shanghai Municipal Science and Technology Commission(16JC1400900)
文摘This paper concerns the problem of object segmentation in real-time for picking system. A region proposal method inspired by human glance based on the convolutional neural network is proposed to select promising regions, allowing more processing is reserved only for these regions. The speed of object segmentation is significantly improved by the region proposal method.By the combination of the region proposal method based on the convolutional neural network and superpixel method, the category and location information can be used to segment objects and image redundancy is significantly reduced. The processing time is reduced considerably by this to achieve the real time. Experiments show that the proposed method can segment the interested target object in real time on an ordinary laptop.
基金the National Natural Science Foundation of China(No.61702323)。
文摘Moving object segmentation (MOS) is one of the essential functions of the vision system of all robots,including medical robots. Deep learning-based MOS methods, especially deep end-to-end MOS methods, are actively investigated in this field. Foreground segmentation networks (FgSegNets) are representative deep end-to-endMOS methods proposed recently. This study explores a new mechanism to improve the spatial feature learningcapability of FgSegNets with relatively few brought parameters. Specifically, we propose an enhanced attention(EA) module, a parallel connection of an attention module and a lightweight enhancement module, with sequentialattention and residual attention as special cases. We also propose integrating EA with FgSegNet_v2 by taking thelightweight convolutional block attention module as the attention module and plugging EA module after the twoMaxpooling layers of the encoder. The derived new model is named FgSegNet_v2 EA. The ablation study verifiesthe effectiveness of the proposed EA module and integration strategy. The results on the CDnet2014 dataset,which depicts human activities and vehicles captured in different scenes, show that FgSegNet_v2 EA outperformsFgSegNet_v2 by 0.08% and 14.5% under the settings of scene dependent evaluation and scene independent evaluation, respectively, which indicates the positive effect of EA on improving spatial feature learning capability ofFgSegNet_v2.
文摘Clustering algorithms in feature space are important methods in image segmentation. The choice of the effective feature parameters and the construction of the clustering method are key problems encountered with clustering algorithms. In this paper, the multifractal dimensions are chosen as the segmentation feature parameters which are extracted from original image and wavelet-transformed image. SOM (Self-Organizing Map) network is applied to cluster the segmentation feature parameters. The experiment shows that the performance of the presented algorithm is very good.
文摘In this paper, we propose several improved neural networks and training strategy using data augmentation to segment human radius accurately and efficiently. This method can provide pixel-level segmentation accuracy through the low-level features of the neural network, and automatically distinguish the classification of radius. The versatility and applicability can be effectively improved by learning and training digital X-ray images obtained from digital X-ray imaging systems of different manufacturers.
文摘neural network model based on backbone propagation was applied to Learn-ing and predicting the interaction between antiparallelly interactive peptide seg-ments in proteins.Hydrophobic properties pf residues were found dominant in in-terpeptides.Weights of each kind of residues, obtained by this work,suggestedsome different scales for the hydrophobicity of the residue.These will be helpful in understanding polypeptide structure and protein folding.
基金Supported by National Science Foundation of China(No.81800878)Interdisciplinary Program of Shanghai Jiao Tong University(No.YG2017QN24)+1 种基金Key Technological Research Projects of Songjiang District(No.18sjkjgg24)Bethune Langmu Ophthalmological Research Fund for Young and Middle-aged People(No.BJ-LM2018002J)
文摘AIM: To explore a segmentation algorithm based on deep learning to achieve accurate diagnosis and treatment of patients with retinal fluid.METHODS: A two-dimensional(2D) fully convolutional network for retinal segmentation was employed. In order to solve the category imbalance in retinal optical coherence tomography(OCT) images, the network parameters and loss function based on the 2D fully convolutional network were modified. For this network, the correlations of corresponding positions among adjacent images in space are ignored. Thus, we proposed a three-dimensional(3D) fully convolutional network for segmentation in the retinal OCT images.RESULTS: The algorithm was evaluated according to segmentation accuracy, Kappa coefficient, and F1 score. For the 3D fully convolutional network proposed in this paper, the overall segmentation accuracy rate is 99.56%, Kappa coefficient is 98.47%, and F1 score of retinal fluid is 95.50%. CONCLUSION: The OCT image segmentation algorithm based on deep learning is primarily founded on the 2D convolutional network. The 3D network architecture proposed in this paper reduces the influence of category imbalance, realizes end-to-end segmentation of volume images, and achieves optimal segmentation results. The segmentation maps are practically the same as the manual annotations of doctors, and can provide doctors with more accurate diagnostic data.
文摘The multilayer feedforward network is used for image segmentation. This paper deals with the procedure of achieving the learning patterns and the method of improving the learning rate. The experiment shows that the image segmentation can get better result from using the multilayer feedforward network.
文摘This paper presents an unsupervised range image segmentation based on Kohonen neural network. At first, the derivative and partial derivative of each point are calculated and the normal in each points is gotten. With the character vectors including normal and range value, self-organization map is introduced to cluster. The normal analysis is used to eliminate over-segmentation and the last result is gotten. This method avoid selecting original seeds and uses fewer samples, moreover computes rapidly. The experiment shows the better performance.
文摘A new approach for multilevel image segmentation based on fuzzy cellular neural network(CNN) is proposed. Based on a novel fuzzy CNN, a new template is proposed for multilevel image segmentation. The result of computer simulation proves this approach is reasonable. The stability of the fuzzy neural network is also analyzed in this paper.
文摘This paper presents a new solution to the image segmentation problem, which is based on fuzzy-neural-network hybrid system (FNNHS). This approach can use the experiential knowledge and the ability of neural networks which learn knowledge from the examples, to obtain the well performed fuzzy rules. Furthermore this fuzzy inference system is completed by neural network structure which can work in parallel. The segmentation process consists of pre-segmentation based on region growing algorithm and region merging based on FNNHS. The experimental results on the complicated image manifest the utility of this method.
基金Supported by American Heart Association Grant in Aid Founders Affiliate No.17GRNT33420119(Mani V)NIH NHLBI 2R01HL070121(Fayad ZA)and NIH NHLBI 1R01HL135878(Fayad ZA)
文摘BACKGROUND Automated,accurate,objective,and quantitative medical image segmentation has remained a challenging goal in computer science since its inception.This study applies the technique of convolutional neural networks(CNNs)to the task of segmenting carotid arteries to aid in the assessment of pathology.AIM To investigate CNN’s utility as an ancillary tool for researchers who require accurate segmentation of carotid vessels.METHODS An expert reader delineated vessel wall boundaries on 4422 axial T2-weighted magnetic resonance images of bilateral carotid arteries from 189 subjects with clinically evident atherosclerotic disease.A portion of this dataset was used to train two CNNs(one to segment the vessel lumen and the other to segment the vessel wall)with the remaining portion used to test the algorithm’s efficacy by comparing CNN segmented images with those of an expert reader.Overall quantitative assessment between automated and manual segmentations was determined by computing the DICE coefficient for each pair of segmented images in the test dataset for each CNN applied.The average DICE coefficient for the test dataset(CNN segmentations compared to expert’s segmentations)was 0.96 for the lumen and 0.87 for the vessel wall.Pearson correlation values and the intra-class correlation coefficient(ICC)were computed for the lumen(Pearson=0.98,ICC=0.98)and vessel wall(Pearson=0.88,ICC=0.86)segmentations.Bland-Altman plots of area measurements for the CNN and expert readers indicate good agreement with a mean bias of 1%-8%.CONCLUSION Although the technique produces reasonable results that are on par with expert human assessments,our application requires human supervision and monitoring to ensure consistent results.We intend to deploy this algorithm as part of a software platform to lessen researchers’workload to more quickly obtain reliable results.
基金supported by the National Natural Science Foundation of China(61073106)the Aerospace Science and Technology Innovation Fund(CASC201105)
文摘An image segmentation algorithm of the restrained fuzzy Kohonen clustering network （RFKCN） based on high- dimension fuzzy character is proposed. The algorithm includes two steps. The first step is the fuzzification of pixels in which two redundant images are built by fuzzy mean value and fuzzy median value. The second step is to construct a three-dimensional （3-D） feature vector of redundant images and their original images and cluster the feature vector through RFKCN, to realize image seg- mentation. The proposed algorithm fully takes into account not only gray distribution information of pixels, but also relevant information and fuzzy information among neighboring pixels in constructing 3- D character space. Based on the combination of competitiveness, redundancy and complementary of the information, the proposed algorithm improves the accuracy of clustering. Theoretical anal- yses and experimental results demonstrate that the proposed algorithm has a good segmentation performance.
文摘<strong>Purpose</strong><span style="font-family:;" "=""><span style="font-family:Verdana;"><strong>: </strong></span><span style="font-family:Verdana;">To improve the liver auto-segmentation performance of three-</span><span style="font-family:Verdana;">dimensional (3D) U-net by replacing the conventional up-sampling convolution layers with the Pixel De-convolutional Network (PDN) that considers spatial features. </span><b><span style="font-family:Verdana;">Methods</span></b><span style="font-family:Verdana;">: The U-net was originally developed to segment neuronal structure with outstanding performance but suffered serious artifacts from indirectly unrelated adjacent pixels in its up-sampling layers. The hypothesis of this study was that the segmentation quality of </span></span><span style="font-family:Verdana;">the </span><span style="font-family:Verdana;">liver could be improved with PDN in which the up-sampling layer was replaced by a pixel de-convolution layer (PDL). Seventy</span><span style="font-family:Verdana;">-</span><span style="font-family:;" "=""><span style="font-family:Verdana;">eight plans of abdominal cancer patients were anonymized and exported. Sixty-two were chosen for training two networks: 1) 3D U-Net, and 2) 3D PDN, by minimizing the Dice loss function. The other sixteen plans were used to test the performance. The similarity Dice and Average Hausdorff Distance (AHD) were calculated and compared between these two networks. </span><b><span style="font-family:Verdana;">Results</span></b><span style="font-family:Verdana;">: The computation time for 62 training cases and 200 training epochs was about 30 minutes for both networks. The segmentation performance was evaluated using the remaining 16 cases. For the Dice score, the mean ± standard deviation were 0.857 ± 0.011 and 0.858 ± 0.015 for the PDN and U-Net, respectively. For the AHD, the mean ± standard deviation were 1.575 ± 0.373 and 1.675 ± 0.769, respectively, corresponding to an improvement of 6.0% and 51.5% of mean and standard deviation for the PDN. </span><b><span style="font-family:Verdana;">Conclusion</span></b><span style="font-family:Verdana;">: The PDN has outperformed the U-Net on liver auto-segmentation. The predicted contours of PDN are more conformal and smoother when compared with</span></span><span style="font-family:Verdana;"> the</span><span style="font-family:Verdana;"> U-Net.</span>
