2017 International Conference on Wavelet Analysis and Pattern Recognition (ICWAPR)
GrabCut algorithm is one of most popular approaches of image segmentation. In practice, the segme... more GrabCut algorithm is one of most popular approaches of image segmentation. In practice, the segmented image using GrabCut algorithm always keeps too much redundant information of background. Therefore, an improved GrabCut algorithm is proposed in this paper which modifies the energy function of GrabCut algorithm. Firstly, the segmented part of object needs to be chosen as foreground by user and is used to gain local prior distribution. Then, the local prior distribution is used to improve energy function to obtain the primary result of image segmentation rather than the traditional iterative one. Finally, the edge detection of the segmented image is further introduced to remove the redundant region. The results of experiments show that the proposed approach resulted in better performance and segmentation than that of the traditional GrabCut.
Multi-objective optimization is a challenging research topic because it involves the simultaneous... more Multi-objective optimization is a challenging research topic because it involves the simultaneous optimization of several complex and conflicting objectives. However multi-objectivity is only one aspect of real-world applications and there is a growing interest in the optimization of solutions that are insensitive to parametric variations as well. A new robust preference multi-objective optimization algorithm is proposed in this paper, and the robust measurement of solution is designed based on Latin Hypercube Sampling, which is embedded in the optimization process to guide the optimization direction and help the better robust solution have more chance to survive. In order to obtain different preference of the robust solutions, a new fitness scheme is also presented. Through the adjustment of fitness function parameter preference, robust solutions can be obtained. Results suggest that the proposed algorithm has a bias towards the region where the preference robust solutions lie.
Lesion detectors based on deep learning can assist doctors in diagnosing diseases. However, the p... more Lesion detectors based on deep learning can assist doctors in diagnosing diseases. However, the performance of current detectors is likely to be unsatisfactory due to the scarcity of training samples. Therefore, it is beneficial to use image generation to augment the training set of a detector. However, when the imaging texture of the medical image is relatively delicate, the synthesized image generated by an existing method may be too poor in quality to meet the training requirements of the detectors. In this regard, a medical image augmentation method, namely, a texture-constrained multichannel progressive generative adversarial network (TMP-GAN), is proposed in this work. TMP-GAN uses joint training of multiple channels to effectively avoid the typical shortcomings of the current generation methods. It also uses an adversarial learning-based texture discrimination loss to further improve the fidelity of the synthesized images. In addition, TMP-GAN employs a progressive generation mechanism to steadily improve the accuracy of the medical image synthesizer. Experiments on the publicly available dataset CBIS-DDMS and our pancreatic tumor dataset show that the precision/recall/F1-score of the detector trained on the TMP-GAN augmented dataset improves by 2.59%/2.70%/2.77% and 2.44%/2.06%/2.36%, respectively, compared to the optimal results of other data augmentation methods. The FROC curve of the detector is also better than the curve from the contrast-augmented trained dataset. Therefore, we believe the proposed TMP-GAN is a practical technique to efficiently implement lesion detection case studies.
2019 IEEE International Conference on Bioinformatics and Biomedicine (BIBM), 2019
Lymphoma detection and segmentation from PET images are critical tasks for cancer staging and tre... more Lymphoma detection and segmentation from PET images are critical tasks for cancer staging and treatment monitoring. However, it is still a challenge owing to the complexities of lymphoma PET data themselves, and the huge memory requirements for 3D volume data. In this work, a prior knowledge integrated scheme based on DBSCAN and GAs is proposed to detect and segment lymphomas in 3D PET images. To reduce memory requirements and add more feature information, billions of voxels in 3D volume data are first aggregated into supervoxels. Then, such supervoxels serve as basic data units for further clustering the supervoxels by using DBSCAN algorithm, in which a new similarity measure based on prior knowledge is proposed. Meanwhile, a genetic algorithm is used to search the most appropriate parameters of DBSCAN to obtain the optimal clustering results. Finally, a series of comparison experiments among various feature attributes and various similarity functions are performed, and the results are encouraging, and show that the proposed scheme by intergrating the prior knowledge of organ distributions can achieve better performance than traditional methods.
Artificial Neural Networks and Machine Learning – ICANN 2018, 2018
Pancreatic cystic neoplasm (PCN) is one of the most common tumors in the digestive tract. It is s... more Pancreatic cystic neoplasm (PCN) is one of the most common tumors in the digestive tract. It is still a challenging task for doctors to diagnose the types of pancreatic cystic neoplasms by using Computed Tomography (CT) images. Especially for serous cystic neoplasms (SCNs) and mucinous cystic neoplasms (MCNs), doctors hardly distinguish one from the other by the naked eyes owing to the high similarities between them. In this work, a multi-channel multiple-classifier (MCMC) model is proposed to distinguish the two pancreatic cystic neoplasms in CT images. At first, multi-channel images are used to enhance the image edge of the tumor, then the residual network is adopted to extract features. Finally, the multiple classifiers are applied to classify the results. Experiments show that the proposed method can effectively improve the classification effect, and the results can help doctors to utilize the CT images to achieve reliable non-invasive disease diagnosis.
2019 IEEE International Conference on Bioinformatics and Biomedicine (BIBM), 2019
Due to the poor resolution and low signal-to-noise ratio in PET images, and especially to the wid... more Due to the poor resolution and low signal-to-noise ratio in PET images, and especially to the wide variation in size, shape, site and SUV value among different patients or even for the same patient, lymphoma segmentation in 3D PET Images is still a challenging task in the field of medical image processing. In this work, a novel non-self background-based data enhancement method is proposed for the deep learning-based lymphoma segmentation problem. Firstly, a lymphoma pool with 1991 lymphoid lesions is created. Then, some lymphomas from the lymphoma pool are randomly selected and integrated into their non-self images of the patients according to their respective coordinates when training networks. Finally, a series of comparison experiments among various network models and methods are conducted to verify the effectiveness of the proposed method. The results indicated that the proposed method was promising, and could obtain better comprehensive performance than other methods without any data enhancements for the lymphoma segmentation problems.
Single image inverse problem is a notoriously challenging ill-posed problem that aims to restore ... more Single image inverse problem is a notoriously challenging ill-posed problem that aims to restore the original image from one of its corrupted versions. Recently, this field has been immensely influenced by the emergence of deep-learning techniques. Deep Image Prior (DIP) offers a new approach that forces the recovered image to be synthesized from a given deep architecture. While DIP is quite an effective unsupervised approach, it is deprecated in real-world applications because of the requirement of human assistance. In this work, we aim to find the best-recovered image without the assistance of humans by adding a stopping criterion, which will reach maximum when the iteration no longer improves the image quality. More specifically, we propose to add a pseudo noise to the corrupted image and measure the pseudo-noise component in the recovered image by the orthogonality between signal and noise. The accuracy of the orthogonal stopping criterion has been demonstrated for several teste...
2020 IEEE International Conference on Bioinformatics and Biomedicine (BIBM), 2020
An activation function is a key component of artificial neural networks (ANNs). It has a great im... more An activation function is a key component of artificial neural networks (ANNs). It has a great impact on the performance and convergence of neural networks. In this work, a self-adapting methodology is proposed to explore the optimal adaptive activation functions for various tasks based on S-shaped or ReLu-shaped activation functions, which are regulated only by introducing several parameters. To verify the effectiveness of the proposed methodology, a series of comparison experiments are performed with MLP, CNN and RNN network structure on the benchmark datasets of image, text and audio. The experimental results are encouraging, and show that the proposed methodology can locate the optimal activation functions for various tasks. Nevertheless, the obtained functions are competitive and the improvements on network performance are significant compared with other popular activation functions, such as ELU, PReLU, ReLU, and Sigmoid.
Shape illustration images (SIIs) are common and important in describing the cross-sections of ind... more Shape illustration images (SIIs) are common and important in describing the cross-sections of industrial products. Same as MNIST, the handwritten digit images, SIIs are gray or binary and containing shapes that are surrounded by large areas of blanks. In this work, Residual-Recursion Autoencoder (RRAE) has been proposed to extract low-dimensional features from SIIs while maintaining reconstruction accuracy as high as possible. RRAE will try to reconstruct the original image several times and recursively fill the latest residual image to the reserved channel of the encoder's input before the next trial of reconstruction. As a kind of neural network training framework, RRAE can wrap over other autoencoders and increase their performance. From experiment results, the reconstruction loss is decreased by 86.47% for convolutional autoencoder with high-resolution SIIs, 10.77% for variational autoencoder and 8.06% for conditional variational autoencoder with MNIST.
With the development of medical imaging technology, an increasing number of images of different m... more With the development of medical imaging technology, an increasing number of images of different modalities are used in diagnosis and fused together for getting more information. Multi-modality image registration is the key technology of multi-modal image fusion. It is also a basic topic in computer vision and widely used in image analysis. However, the images of different modalities would not show the same representation. Therefore, images of different modalities cannot be aligned directly with each other. Currently, a new descriptor which is calculated from different modality images by self-similarities context has been proposed. This descriptor makes images of different modalities align with each other directly. However, it is still affect by noise. In this paper, a new framework of multimodal registration is proposed to improve the method of multimodal image registration with self-similarities contexts. Firstly, a series of operations of de-noise is implemented in images before r...
2010 3rd International Conference on Computer Science and Information Technology, 2010
In the past decades, the main focus of optimization studies including multi-objective optimizatio... more In the past decades, the main focus of optimization studies including multi-objective optimization was placed on finding the global optimum or Pareto-optimal solutions. However, in practical engineering, users may not always be interested in finding the so-called global best solutions, particularly when these Pareto-optimal solutions are quite sensitive to the variable perturbations, which cannot be avoided in practice. In such
2017 International Conference on Wavelet Analysis and Pattern Recognition (ICWAPR)
GrabCut algorithm is one of most popular approaches of image segmentation. In practice, the segme... more GrabCut algorithm is one of most popular approaches of image segmentation. In practice, the segmented image using GrabCut algorithm always keeps too much redundant information of background. Therefore, an improved GrabCut algorithm is proposed in this paper which modifies the energy function of GrabCut algorithm. Firstly, the segmented part of object needs to be chosen as foreground by user and is used to gain local prior distribution. Then, the local prior distribution is used to improve energy function to obtain the primary result of image segmentation rather than the traditional iterative one. Finally, the edge detection of the segmented image is further introduced to remove the redundant region. The results of experiments show that the proposed approach resulted in better performance and segmentation than that of the traditional GrabCut.
Multi-objective optimization is a challenging research topic because it involves the simultaneous... more Multi-objective optimization is a challenging research topic because it involves the simultaneous optimization of several complex and conflicting objectives. However multi-objectivity is only one aspect of real-world applications and there is a growing interest in the optimization of solutions that are insensitive to parametric variations as well. A new robust preference multi-objective optimization algorithm is proposed in this paper, and the robust measurement of solution is designed based on Latin Hypercube Sampling, which is embedded in the optimization process to guide the optimization direction and help the better robust solution have more chance to survive. In order to obtain different preference of the robust solutions, a new fitness scheme is also presented. Through the adjustment of fitness function parameter preference, robust solutions can be obtained. Results suggest that the proposed algorithm has a bias towards the region where the preference robust solutions lie.
Lesion detectors based on deep learning can assist doctors in diagnosing diseases. However, the p... more Lesion detectors based on deep learning can assist doctors in diagnosing diseases. However, the performance of current detectors is likely to be unsatisfactory due to the scarcity of training samples. Therefore, it is beneficial to use image generation to augment the training set of a detector. However, when the imaging texture of the medical image is relatively delicate, the synthesized image generated by an existing method may be too poor in quality to meet the training requirements of the detectors. In this regard, a medical image augmentation method, namely, a texture-constrained multichannel progressive generative adversarial network (TMP-GAN), is proposed in this work. TMP-GAN uses joint training of multiple channels to effectively avoid the typical shortcomings of the current generation methods. It also uses an adversarial learning-based texture discrimination loss to further improve the fidelity of the synthesized images. In addition, TMP-GAN employs a progressive generation mechanism to steadily improve the accuracy of the medical image synthesizer. Experiments on the publicly available dataset CBIS-DDMS and our pancreatic tumor dataset show that the precision/recall/F1-score of the detector trained on the TMP-GAN augmented dataset improves by 2.59%/2.70%/2.77% and 2.44%/2.06%/2.36%, respectively, compared to the optimal results of other data augmentation methods. The FROC curve of the detector is also better than the curve from the contrast-augmented trained dataset. Therefore, we believe the proposed TMP-GAN is a practical technique to efficiently implement lesion detection case studies.
2019 IEEE International Conference on Bioinformatics and Biomedicine (BIBM), 2019
Lymphoma detection and segmentation from PET images are critical tasks for cancer staging and tre... more Lymphoma detection and segmentation from PET images are critical tasks for cancer staging and treatment monitoring. However, it is still a challenge owing to the complexities of lymphoma PET data themselves, and the huge memory requirements for 3D volume data. In this work, a prior knowledge integrated scheme based on DBSCAN and GAs is proposed to detect and segment lymphomas in 3D PET images. To reduce memory requirements and add more feature information, billions of voxels in 3D volume data are first aggregated into supervoxels. Then, such supervoxels serve as basic data units for further clustering the supervoxels by using DBSCAN algorithm, in which a new similarity measure based on prior knowledge is proposed. Meanwhile, a genetic algorithm is used to search the most appropriate parameters of DBSCAN to obtain the optimal clustering results. Finally, a series of comparison experiments among various feature attributes and various similarity functions are performed, and the results are encouraging, and show that the proposed scheme by intergrating the prior knowledge of organ distributions can achieve better performance than traditional methods.
Artificial Neural Networks and Machine Learning – ICANN 2018, 2018
Pancreatic cystic neoplasm (PCN) is one of the most common tumors in the digestive tract. It is s... more Pancreatic cystic neoplasm (PCN) is one of the most common tumors in the digestive tract. It is still a challenging task for doctors to diagnose the types of pancreatic cystic neoplasms by using Computed Tomography (CT) images. Especially for serous cystic neoplasms (SCNs) and mucinous cystic neoplasms (MCNs), doctors hardly distinguish one from the other by the naked eyes owing to the high similarities between them. In this work, a multi-channel multiple-classifier (MCMC) model is proposed to distinguish the two pancreatic cystic neoplasms in CT images. At first, multi-channel images are used to enhance the image edge of the tumor, then the residual network is adopted to extract features. Finally, the multiple classifiers are applied to classify the results. Experiments show that the proposed method can effectively improve the classification effect, and the results can help doctors to utilize the CT images to achieve reliable non-invasive disease diagnosis.
2019 IEEE International Conference on Bioinformatics and Biomedicine (BIBM), 2019
Due to the poor resolution and low signal-to-noise ratio in PET images, and especially to the wid... more Due to the poor resolution and low signal-to-noise ratio in PET images, and especially to the wide variation in size, shape, site and SUV value among different patients or even for the same patient, lymphoma segmentation in 3D PET Images is still a challenging task in the field of medical image processing. In this work, a novel non-self background-based data enhancement method is proposed for the deep learning-based lymphoma segmentation problem. Firstly, a lymphoma pool with 1991 lymphoid lesions is created. Then, some lymphomas from the lymphoma pool are randomly selected and integrated into their non-self images of the patients according to their respective coordinates when training networks. Finally, a series of comparison experiments among various network models and methods are conducted to verify the effectiveness of the proposed method. The results indicated that the proposed method was promising, and could obtain better comprehensive performance than other methods without any data enhancements for the lymphoma segmentation problems.
Single image inverse problem is a notoriously challenging ill-posed problem that aims to restore ... more Single image inverse problem is a notoriously challenging ill-posed problem that aims to restore the original image from one of its corrupted versions. Recently, this field has been immensely influenced by the emergence of deep-learning techniques. Deep Image Prior (DIP) offers a new approach that forces the recovered image to be synthesized from a given deep architecture. While DIP is quite an effective unsupervised approach, it is deprecated in real-world applications because of the requirement of human assistance. In this work, we aim to find the best-recovered image without the assistance of humans by adding a stopping criterion, which will reach maximum when the iteration no longer improves the image quality. More specifically, we propose to add a pseudo noise to the corrupted image and measure the pseudo-noise component in the recovered image by the orthogonality between signal and noise. The accuracy of the orthogonal stopping criterion has been demonstrated for several teste...
2020 IEEE International Conference on Bioinformatics and Biomedicine (BIBM), 2020
An activation function is a key component of artificial neural networks (ANNs). It has a great im... more An activation function is a key component of artificial neural networks (ANNs). It has a great impact on the performance and convergence of neural networks. In this work, a self-adapting methodology is proposed to explore the optimal adaptive activation functions for various tasks based on S-shaped or ReLu-shaped activation functions, which are regulated only by introducing several parameters. To verify the effectiveness of the proposed methodology, a series of comparison experiments are performed with MLP, CNN and RNN network structure on the benchmark datasets of image, text and audio. The experimental results are encouraging, and show that the proposed methodology can locate the optimal activation functions for various tasks. Nevertheless, the obtained functions are competitive and the improvements on network performance are significant compared with other popular activation functions, such as ELU, PReLU, ReLU, and Sigmoid.
Shape illustration images (SIIs) are common and important in describing the cross-sections of ind... more Shape illustration images (SIIs) are common and important in describing the cross-sections of industrial products. Same as MNIST, the handwritten digit images, SIIs are gray or binary and containing shapes that are surrounded by large areas of blanks. In this work, Residual-Recursion Autoencoder (RRAE) has been proposed to extract low-dimensional features from SIIs while maintaining reconstruction accuracy as high as possible. RRAE will try to reconstruct the original image several times and recursively fill the latest residual image to the reserved channel of the encoder's input before the next trial of reconstruction. As a kind of neural network training framework, RRAE can wrap over other autoencoders and increase their performance. From experiment results, the reconstruction loss is decreased by 86.47% for convolutional autoencoder with high-resolution SIIs, 10.77% for variational autoencoder and 8.06% for conditional variational autoencoder with MNIST.
With the development of medical imaging technology, an increasing number of images of different m... more With the development of medical imaging technology, an increasing number of images of different modalities are used in diagnosis and fused together for getting more information. Multi-modality image registration is the key technology of multi-modal image fusion. It is also a basic topic in computer vision and widely used in image analysis. However, the images of different modalities would not show the same representation. Therefore, images of different modalities cannot be aligned directly with each other. Currently, a new descriptor which is calculated from different modality images by self-similarities context has been proposed. This descriptor makes images of different modalities align with each other directly. However, it is still affect by noise. In this paper, a new framework of multimodal registration is proposed to improve the method of multimodal image registration with self-similarities contexts. Firstly, a series of operations of de-noise is implemented in images before r...
2010 3rd International Conference on Computer Science and Information Technology, 2010
In the past decades, the main focus of optimization studies including multi-objective optimizatio... more In the past decades, the main focus of optimization studies including multi-objective optimization was placed on finding the global optimum or Pareto-optimal solutions. However, in practical engineering, users may not always be interested in finding the so-called global best solutions, particularly when these Pareto-optimal solutions are quite sensitive to the variable perturbations, which cannot be avoided in practice. In such
Uploads