While Vision-Language Models (VLMs) have garnered increasing attention in the AI community due to their promising practical applications, they exhibit persistent hallucination issues, generating outputs misaligned with visual inputs. Recent studies attribute these hallucinations to VLMs' over-reliance on linguistic priors and insufficient visual feature integration, proposing heuristic decoding calibration strategies to mitigate them. However, the non-trainable nature of these strategies inherently limits their optimization potential. To this end, we propose an adversarial parametric editing framework for Hallucination mitigation in VLMs, which follows an Activate-Locate-Edit Adversarially paradigm. Specifically, we first construct an activation dataset that comprises grounded responses (positive samples attentively anchored in visual features) and hallucinatory responses (negative samples reflecting LLM prior bias and internal knowledge artifacts). Next, we identify critical hallucination-prone parameter clusters by analyzing differential hidden states of response pairs. Then, these clusters are fine-tuned using prompts injected with adversarial prefixes optimized via prompt tuning to maximize visual neglect, thereby forcing the model to prioritize visual evidence over inherent parametric biases. Evaluations on both generative and discriminative VLM tasks demonstrate the significant effectiveness of ALEAHallu in alleviating hallucinations.

Semantic segmentation models have made significant progress by optimizing for the characteristics of medical images, but their accuracy and generalization ability still need improvement. The vision foundation model SAM excels in segmentation accuracy and generalization, but its performance drops significantly when directly applied to medical image segmentation due to its training on natural images. To address this, we propose a distillation framework with SAM as the teacher and medical segmentation model as the student, distilling SAM’s powerful feature extraction capabilities into medical segmentation model to enhance its performance. Directly using existing knowledge distillation methods introduces two challenges: 1) a large semantic gap between SAM and medical segmentation model’s feature layers; 2) SAM’s strong edge feature extraction ability is underutilized. In this paper, we present the SAM-Based Knowledge Distillation with Adaptive Fusion and Edge Features (SAM-AEKD), consisting of two modules: FAFM and EFEM. FAFM dynamically weights and convolves multi-layer features from the teacher or student encoders, avoiding semantic mismatches during distillation. EFEM extracts edge features from the fused features using multi-scale convolution and channel aggregation, enabling knowledge distillation in both edge and fused feature spaces, enhancing the student’s edge feature extraction ability. Experiments on three medical modality datasets show that SAM-AEKD significantly improves students’ segmentation accuracy, outperforming other distillation methods.

Medical image segmentation has improved with advances in model architectures and feature extraction, yet current SOTA models still face challenges in generalization and accuracy. The vision foundation model SAM demonstrates strong segmentation precision and zero-shot generalization on natural images. However, due to domain gaps, SAM, even when fine-tuned or used as a backbone, underperforms compared to medical SOTA models. To address this, we propose a knowledge distillation framework that treats SAM as a teacher to transfer its powerful feature extraction and generalization abilities to a medical segmentation model, thereby improving both performance and domain robustness. Nevertheless, in this setting, existing distillation methods face two key challenges: (1) Existing feature-level knowledge distillation methods either rely on subjective assumptions for feature alignment or use attention-based mechanisms to align teacher-student features, which are computationally expensive. Moreover, since recent SOTA models in medical image segmentation are primarily Transformer-based, research on distilling SAM’s feature extraction capabilities into Trans-former-based medical models remains scarce. (2) SAM has strong edge perception, but existing distillation methods focus on global alignment, leaving its edge-aware capabilities largely underutilized. To tackle these issues, we propose the SAM-Teacher framework, which consists of Adaptive Feature Alignment Module (AFAM) and Edge Perception Distillation Module (EPDM). AFAM leverages gradients from segmentation loss to adaptively determine the alignment direction between teacher and student features, and incorporates an attention-based distillation loss tailored to Transformer models. EDPM explicitly strengthens the distillation of edge information, addressing the limitation of existing methods that treat edge and non-edge regions equally. Extensive experiments on three modal medical image datasets show that SAM-Teacher significantly improves the segmentation accuracy and generalization of the student model, outperforming existing distillation methods. Ablation studies further validate the effectiveness of each module.

This is a corrigendum for the article “An Unbiased Risk Estimator for Partial Label Learning with Augmented Classes” published in ACM Trans. Intell. Syst. Technol. 15(6): 131:1-131:22 (2024).

Segment anything model (SAM) has shown promising segmentation capabilities , however its performance significantly declines when applied to ultrasound images. Many works have emerged to address this issue, but still have two deficiencies: 1) SAM fine-tuning relies on manual prompts, which only allowing semiautomatics segmentation; 2) They exclusively utilize the frozen SAM encoder as the sole image encoder, which lacks pathological information. In this paper, we propose US-SAM which improve SAM with three modules: Pathological Extractor (PE), Fusion Module (FM), and Automatic Prompt Module (APM). PE extracts semantic information of lesions within the ultrasound image. Furthermore, FM fuses the features from PE and SAM image encoder. Finally, APM automatically generates prompts required by SAM using the fused features and uncertainty map. Through extensive experiments on two public ultrasound datasets BUSI and TN3k, our method outperforms other medical SAM methods by nearly 17% in Dice and IOU scores without any prompt from human.

The EGFR mutation status significantly influences targeted therapy for non-small cell lung cancer. In recent years, there has been significant progress in non-invasive EGFR mutation status prediction studies based on chest CT images. However, these studies commonly rely on extensive private data for training rather than small-scale publicly available dataset, thereby failing to overcome the dependency on high-cost large-scale annotated data. Additionally, these studies generally neglect the texture and contour features with strong discriminative power, leading to insufficient performance. This paper proposes a two-stage framework for EGFR mutation status prediction. Initially, we utilize self-supervised Masked Autoencoders (MAE) to pre-train the encoder on in-domain chest CT images, overcoming the problem of insufficient annotated data to reduce the dependency on high-cost annotated data. Subsequently, fine-tune the encoder to make it suitable for downstream EGFR prediction. Simultaneously, we propose texture and contour enhanced MAE (TCMAE), designing Multi-layer Features Aggregation Module (MFAM) to fully exploit multi-layer semantic features, introducing Texture and Contour Prediction Module (TCPM) to enhance the model’s capability in extracting texture and contour features through multitask learning, utilizing Modified Spectral Block (MSB) to adjust the weighting between high and low frequency features. Experiments demonstrate that despite using only a small-scale public dataset, NSCLC-Radiogenomics, the proposed method still achieves high accuracy.

Chest X-ray is the most commonly used detection method for lung diseases, but manual screening often has omissions, so computer-aided diagnosis of chest X-ray abnormalities is necessary. However, since abnormal data relies on expert annotation and is difficult to obtain, unsupervised anomaly detection (UAD) using only normal data has become the focus of attention in the field of medical images. Current UAD methods based on reconstruction often use reconstruction error of original image and reconstructed image as the anomaly score, but the strong reconstruction ability of the autoencoder results in small abnormal image reconstruction error, which is similar to the normal image reconstruction error, making the detection results not satisfactory. Therefore, we proposed CGMAE. In the training stage, masked autoencoder is used as the generator to reconstruct the image, and a discriminator with the reconstructed image and the original image as input is added at the end. Through adversarial learning, the discriminator learns the distribution of normal data. In the testing stage, Gaussian distribution is used to construct the anomaly score, increasing the gap between normal and abnormal data, which is conducive to separating abnormal data from normal data. At the same time, it was found that the current reconstruction models for chest X-ray anomaly detection did not consider the different importance of chest X-ray foreground and background when reconstructing images better. Therefore, we designed a regional weighted loss to enable the generator to reconstruct high-resolution chest X-ray images and enhance the discriminator’s ability to learn data distribution. Experiments on two public datasets, Zhanglab dataset and Chexpert dataset, show that CGMAE exceeds SOTA by 1.41% and 0.98% in AUC metrics respectively.

Currently, anomaly detection methods for retinal OCT images can be mainly divided into two types: reconstruction-based and knowledge distillation-based. The former has weak inter-class descriptiveness and unclear feature boundaries as it is trained with normal images only, and without comparing with abnormal images. The latter is insensitive to rare diseases and lacks a pretrained model with high-performance. In previous works, these two methods are used independently, thus unable to address the innate limitations of the model effectively. Therefore, we proposed a dual-branch model named DRTNet_AD. The two branches are connected through a student network (Encoder) serving as an intermediary bridge. The knowledge distillation network branch learns descriptive features through a pretrained model, while the reconstruction network branch enhances the model’s sensitivity to rare diseases by learning the distribution of normal images. Furthermore, in order to obtain a pretrained model with good feature extraction and descriptive abilities, we proposed a strategy to construct pseudo-anomaly and a layer structure context-aware module (LCAM). DRTNet_AD achieves the SOTA performance on public dataset SpectralisOCT with AUC scores of 98.25%

The key of object goal visual navigation is to learn the spatial relationships between environmental objects and assess their semantic correlations with the target object. We propose an end-to-end visual navigation model based on deep reinforcement learning, called G2SNet, which consists of two feature maps and a specialized fusion feature network: GloVe Feature Map (GFM), Sbbox Feature Map (SFM), and GloVe fusion Network (GNet). GFM represents the position and the semantic information of the objects contained in the observation image, which addresses the issue of the interference in the target recognition caused by the complex background information in the observed image. SFM provides the object sizes in the field of view to assist the distance judgment. GNet relies entirely on network learning to compute the semantic correlations and spatial positional relationships among objects in the environment, enabling the agent to possess better generalization capabilities. This allows learning of spatial relationships between objects in GFM. Experiments on AI2-THOR demonstrate the effectiveness of our proposed three new structures, and the average SPL of the four known scenarios is increased by 16.6%.

Partial Label Learning (PLL) is a typical weakly supervised learning task, which assumes each training instance is annotated with a set of candidate labels containing the ground-truth label. Recent PLL methods adopt identification-based disambiguation to alleviate the influence of false positive labels and achieve promising performance. However, they require all classes in the test set to have appeared in the training set, ignoring the fact that new classes will keep emerging in real applications. To address this issue, in this article, we focus on the problem of Partial Label Learning with Augmented Class (PLLAC), where one or more augmented classes are not visible in the training stage but appear in the inference stage. Specifically, we propose an unbiased risk estimator with theoretical guarantees for PLLAC, which estimates the distribution of augmented classes by differentiating the distribution of known classes from unlabeled data and can be equipped with arbitrary PLL loss functions. Besides, we provide a theoretical analysis of the estimation error bound of the estimator, which guarantees the convergence of the empirical risk minimizer to the true risk minimizer as the number of training data tends to infinity. Furthermore, we add a risk-penalty regularization term in the optimization objective to alleviate the influence of the over-fitting issue caused by negative empirical risk. Extensive experiments on benchmark, UCI, and real-world datasets demonstrate the effectiveness of the proposed approach.

The emergence of Graph Neural Networks (GNNs) has greatly advanced the development of recommendation systems. Recently, many researchers have leveraged GNN-based models to learn fair representations for users and items. However, current GNN-based models suffer from biased user–item interaction data, which negatively impacts recommendation fairness. Although there have been several studies employing adversarial learning to mitigate this issue in recommendation systems, they mostly focus on modifying the model training approach with fairness regularization and neglect direct intervention of biased interaction. In contrast to these models, this article introduces a novel perspective by directly intervening in observed interactions to generate a counterfactual graph (called FairGap) that is not influenced by sensitive node attributes, enabling us to learn fair representations for users and items easily. We design FairGap to answer the key counterfactual question: “Would interactions with an item remain unchanged if a user’s sensitive attributes were concealed?”. We also provide theoretical proofs to show that our learning strategy via the counterfactual graph is unbiased in expectation. Moreover, we propose a fairness-enhancing mechanism to continuously improve user fairness in the graph-based recommendation. Extensive experimental results against state-of-the-art competitors and base models on three real-world datasets validate the effectiveness of our proposed model.

Recently, some span-based methods have achieved encouraging performances for joint aspect-sentiment analysis, which first extract aspects (aspect extraction) by detecting aspect boundaries and then classify the span-level sentiments (sentiment classification). However, most existing approaches either sequentially extract task-specific features, leading to insufficient feature interactions, or they encode aspect features and sentiment features in a parallel manner, implying that feature representation in each task is largely independent of each other except for input sharing. Both of them ignore the internal correlations between the aspect extraction and sentiment classification. To solve this problem, we novelly propose a hierarchical interactive network (HI-ASA) to model two-way interactions between two tasks appropriately, where the hierarchical interactions involve two steps: shallow-level interaction and deep-level interaction. First, we utilize cross-stitch mechanism to combine the different task-specific features selectively as the input to ensure proper two-way interactions. Second, the mutual information technique is applied to mutually constrain learning between two tasks in the output layer, thus the aspect input and the sentiment input are capable of encoding features of the other task via backpropagation. Extensive experiments on three real-world datasets demonstrate HI-ASA’s superiority over baselines.

High-resolution (HR) magnetic resonance (MR) images could provide reliable visual information for clinical diagnosis. Recently, super-resolution (SR) methods based on convolutional neural networks (CNNs) have shown great potential in obtaining HR MR images. However, most existing CNN-based SR methods neglect the internal priors of the MR image, which hides the performance of SR. In this work, we propose a 3D cross-scale feature transformer network (CFTN) to utilize the cross-scale priors within MR features. Specifically, we stack multiple 3D residual channel attention blocks (RCABs) as the backbone. Meanwhile, we design a plug-in mutual-projection feature enhancement module (MFEM) to extract the target-scale features with HR cues, which is able to capture the global cross-scale self-similarity within features and can be flexibly inserted into any position of the backbone. Furthermore, we propose a spatial attention fusion module (SAFM) to adaptively adjust and fuse the target-scale features and up-sampled features that are respectively extracted by the MFEM and the backbone. Experimental results show that our CFTN achieves a new state-of-the-art MR image SR performance.

Low-dose computed tomography (CT) has attracted widespread attention in the medical imaging field due to its mild radiation hazards to the human body. However, the image may suffer from unpleasing noises under the low-dose radiation condition, which is not conducive to accurate analysis and diagnosis of diseases. Recently, deep learning has shown great potential in low-dose CT denoising. However, current approaches neglect that noise causes varying degrees of damage to low-/high-frequency components of the LDCT image, which hinders further improvement in denoising accuracy. In this paper, we propose a novel frequency separation network (FSNet) for low-dose CT image denoising, which recovers low-/high-frequency components separately, and takes full advantage of them for reconstructing high-quality CT images. To recover clean frequency components effectively, we design hybrid convolution module (HCM) that exploits parallel cascaded convolution path and encoder-decoder path to eliminate noises and preserve image structures. To fuse clean frequency components adaptively, we introduce content attention module (CAM) to adjust the contribution of features across valuable channels and regions, which encourages FSNet to restore image contents based on their frequency characteristics. Extensive experimental results on the Mayo Clinic low-dose CT image dataset show that our proposed FSNet outperforms state-of-the-art denoising methods.

High resolution (HR) magnetic resonance (MR) images are crucial for medical diagnosis. However, in practice, low resolution MR images are often acquired due to hardware limitation. In this work, we propose a gating feature dense network to reconstruct HR MR images from low resolution acquisitions, where we use local residual dense block (LRDB) as the backbone. We propose gating mechanism, which includes absorption gate and release gate, to adaptively introduce the informative features of previous LRDBs to current LRDB to solve the problem of insufficient features sharing. The absorption gate can fuse the output feature of LRDBs with adaptive weights, which allows the model to adaptively learn the effects of different LRDBs for MR image super-resolution (SR). Experimental results show that our proposed method achieves a new state-of-the-art quantitative and visual performance in anisotropic MR image SR.

Pancreas segmentation has been challenging due to its large variations in size, shape, localization, and indistinguishable boundary. Current mainstream pancreas segmentation methods are based on the deep encoder-decoder structure, which recover high-resolution representations from encoded low-resolution representations to generate pancreas voxel masks. However, the details of the pancreas are easily lost in the encoding stage. In this paper, we propose a 3D high-resolution network (3D HRNet) to extract pancreas features, which maintains high-resolution representations throughout the whole process. We use a novel recurrent gated fusion (RGF) head to fuse high-resolution features and generate pancreas voxel masks. To reduce variable background interference, we crop the pancreas area from abdominal CT images for segmentation with a pancreas localization network. We evaluate the above propsed method on the public NIH and MSD pancreas segmentation datasets, and experiments show a competitive result with a mean Dice-Srensen Coefficient (DSC) of 85.82±4.01% on NIH and 84.22±5.91% on MSD, respectively. The lowest variance and the highest mean DSC reveal the stability of our method among current methods and its potential in the clinical setting.

Pulmonary nodule detection in low-dose computed tomography (CT) images is essential for early screening and treatment of lung cancer. Previous related researches based on deep convolutional neural networks generally rely on 2D or 2.5D components and only focus on the output feature information under a single receptive field. Considering the 3D nature of lung CT images and the performance limitation of state-of-the-art nodule detection methods, we develop a novel 3D multi-branch region proposal network with an encoder-decoder structure. Specifically, each parallel branch is designed with 3D residual blocks and U-Net-like structure to effectively extract multi-scale fusion features based on 3D spatial information of CT scans, and the strategies of varying receptive fields and sharing weight parameters are used to improve the sensitivity of the detection network to nodules with scale variation and maintain the original parameters. Besides, we propose a multi-scale attentional feature fusion module to better fuse high-resolution and semantically strong features and adaptively learn the inter-dependency information of different feature maps. Finally, we compare a dynamically scaled cross entropy loss and online hard example mining (OHEM) to combat the imbalance of positive and negative samples during training, which is aimed at assisting with network optimization. Our extensive experiments on publicly available CT scans obtained from LUNA16 and TianChi1competition dataset demonstrate that our method outperform state-of-the-art pulmonary nodule detection models.

Search and inference are two main strategies for optimally solving Distributed Constraint Optimization Problems (DCOPs). Recently, several algorithms were proposed to combine their advantages. Unfortunately, such algorithms only use an approximated inference as a one-shot preprocessing phase to construct the initial lower bounds which lead to inefficient pruning under the limited memory budget. On the other hand, iterative inference algorithms (e.g., MB-DPOP) perform a context-based complete inference for all possible contexts but suffer from tremendous traffic overheads. In this paper, (i) hybridizing search with context-based inference, we propose a complete algorithm for DCOPs, named HS-CAI where the inference utilizes the contexts derived from the search process to establish tight lower bounds while the search uses such bounds for efficient pruning and thereby reduces contexts for the inference. Furthermore, (ii) we introduce a context evaluation mechanism to select the context patterns for the inference to further reduce the overheads incurred by iterative inferences. Finally, (iii) we prove the correctness of our algorithm and the experimental results demonstrate its superiority over the state-of-the-art.

Magnetic resonance images (MRI) with high spatial resolution provide detailed anatomical information for accurate disease diagnosis and quantitative analysis. However, the resolution of clinical MRI is restricted by hardware limitation and cost. Recently, convolutional neural networks (CNNs) are utilized to improve the spatial resolution of MRI. Whereas, current CNNbased super-resolution (SR) methods treat different types and levels of features equally, which hinders the representation ability of network. In this paper, we propose a novel feature attention super-resolution (FASR) network to adaptively capture different informative features. FASR uses parallel channel and spatial attention to enhance valuable features and suppress redundant information. The refined features are upsampled using a subpixel convolution layer, and then fused to predict the missing high-resolution details. To accelerate the training and generate realistic MRI, we introduce cross-scale residual and adversarial training to train FASR. Experimental results on brain MRI datasets of healthy subjects and gliomas show that the proposed FASR achieves a new state-of-the-art MRI SR performance.

We propose a novel model-based network to reconstruct the magnetic resonance (MR) image. In this network, the Approximate Message Passing (AMP) algorithm is unrolled to solve the optimization problem of compressed sensing MR imaging, and several CNN blocks is embedded as de-aliasing steps. We relax the restriction on the parameter selection of AMP algorithm, and enable the parameters trainable in our proposed method. Each CNN block and AMP block is followed by a data consistency (DC) operation, which can efficiently accelerate the convergence of the reconstruction network. The trainable parameters of our DC share the weights and can automatically adapt to the error pattern. Experimental results show that the proposed method obtains faster convergence speed and achieves a new state-of-the-art MR image reconstruction performance.

Belief propagation approaches, such as Max-Sum and its variants, are important methods to solve large-scale Distributed Constraint Optimization Problems (DCOPs). However, for problems with n-ary constraints, these algorithms face a huge challenge since their computational complexity scales exponentially with the number of variables a function holds. In this paper, we present a generic and easy-touse method based on a branch-and-bound technique to solve the issue, called Function Decomposing and State Pruning (FDSP). We theoretically prove that FDSP can provide monotonically non-increasing upper bounds and speed up belief propagation based incomplete DCOP algorithms without an effect on solution quality. Also, our empirically evaluation indicates that FDSP can reduce 97% of the search space at least and effectively accelerate Max-Sum, compared with the state-of-the-art.

High resolution (HR) magnetic resonance images (MRI) provide rich tissue anatomical information that enables accurate diagnostics and pathological analysis. However, the acquisition of HR MRI is limited by hardware restrictions, scanning time, and signal-to-noise ratio (SNR) in clinical applications. Recently, deep learning has shown promising power for improving the spatial resolution of MRI. In this study, we propose a multilevel and parallel Conv-Deconv super-resolution (CDSR) network to reconstruct high-quality HR MRI from low resolution (LR) inputs. Different from current SR methods based on convolutional neural networks (CNNs), we connect parallel 3D convolution and deconvolution filters to capture context information and extract multi-level features. Hierarchical features are adaptively upsampled using each of their following deconvolution layers and then fused together to recover the HR details. In order to alleviate the optimization difficulty, we introduce the interpolated input to the fused output, which performs like a cross-scale residual learning strategy, hence accelerates the convergence. Extensive experimental results on three benchmark datasets show that our proposed method outperforms current reported MRI SR methods and sets a new state-of-the-art performance.

In the natural language environment, accurately inferring the meaning of a token according to its context is crucial to understanding a sophisticated expression. However, this is not easy for a machine. The traditional language models used to train distributed word vectors are often restricted by single-sense embedding. In this paper, we develop a model called MSCvec (Multi-sense Soft Clustering Vector) for word sense disambiguation of polysemy in context. We extract the features of individual words by the co-occurrence PPMI (Positive Pointwise Mutual Information) matrix, and decompose the matrix by NMF (Nonnegative Matrix Factorization) into low-rank representations of target words, which are used as the input of an unsupervised sparse soft clustering method called Sparse Fuzzy C-means (SFCM). We use SFCM to determine the global semantic space of words, and partition the subspaces of multiple senses of a polysemous word. We relabel candidate words by the negative average log likelihood, and train multi-sense embedding with extensional vocabulary by the fastText model. Compared with the traditional static embeddings, the result shows that NMF and SFCM design can improve the performance in word similarity and relatedness tasks as well as in text classification tasks of different types of text. Accurate semantic representation of MSCvec would be necessary to produce outstanding results.

Because of affected by weather conditions, camera pose and range, etc. objects are usually small, blurry, occluded and diverse pose in the images, which are gathered from outdoor surveillance cameras or access control system. It is challenging and important to detect faces precisely for face recognition system in the field of public security. In this paper, we design a context modeling network named Feature Hierarchy Encoder-Decoder Network for face detection (FHEDN), which can detect small, blurry and occluded faces hierarchy by hierarchy from the end to the beginning in a single stage. The proposed network consists of encoder and decoder subnetworks. The encoder subnetwork constructs a multi-scale feature hierarchy pyramid through VGG-16 as backbone network. The decoder subnetwork models context semantic information around face and fuses it into the feature hierarchy for face detection. In addition, we analyze the influence of distribution of training set, scale of feature hierarchy and receipt field size on the detection performance in implement stage. The experiments demonstrate that our network achieves the promising performance on AFW, PASCAL FACE, WIDER FACE and FDDB benchmarks.

In magnetic resonance (MR), hardware limitations, scan time constraints, and patient movement often result in the acquisition of anisotropic 3-D MR images with limited spatial resolution in the out-of-plane views. Our goal is to construct an isotropic high-resolution (HR) 3-D MR image through upsampling and fusion of orthogonal anisotropic input scans. We propose a multiframe super-resolution (SR) reconstruction technique based on sparse representation of MR images. Our proposed algorithm exploits the correspondence between the HR slices and the low-resolution (LR) sections of the orthogonal input scans as well as the self-similarity of each input scan to train pairs of overcomplete dictionaries that are used in a sparse-land local model to upsample the input scans. The upsampled images are then combined using wavelet fusion and error backprojection to reconstruct an image. Features are learned from the data and no extra training set is needed. Qualitative and quantitative analyses were conducted to evaluate the proposed algorithm using simulated and clinical MR scans. Experimental results show that the proposed algorithm achieves promising results in terms of peak signal-to-noise ratio, structural similarity image index, intensity profiles, and visualization of small structures obscured in the LR imaging process due to partial volume effects. Our novel SR algorithm outperforms the nonlocal means (NLM) method using self-similarity, NLM method using self-similarity and image prior, self-training dictionary learning-based SR method, averaging of upsampled scans, and the wavelet fusion method. Our SR algorithm can reduce through-plane partial volume artifact by combining multiple orthogonal MR scans, and thus can potentially improve medical image analysis, research, and clinical diagnosis.

The RANSAC is widely used in image registration algorithms. However, the RANSAC becomes computationally expensive when the number of feature points is large. And also, its high error-matching ratio caused by the large number of iterations always raises the possibility of false registration. To deal with these drawbacks, a novel multi-objective optimization-based image registration method is proposed, named MO-IRM. In MO-IRM, a multi-objective estimation model is built to describe the feature matching pairs (data set), with no need for the pre-check process that is necessary in some improved RANSAC algorithms to eliminate the error-matching pairs. Moreover, a full variate Gaussian model-based RM-MEDA without clustering process (FRM-MEDA) is presented to solve the established multi-objective model. FRM-MEDA only requires a few iterations to find out a correct model. FRM-MEDA can not only greatly reduce the computational overhead but also effectively decrease the possibility of false registration. The proposed MO-IRM is compared with RM-MEDA, NSGA- and the RANSAC based registration algorithm on the Dazu grottoes image database. The experiment results demonstrate that the proposed method achieves ideal registration performances on both two images and multiple images, and greatly outperforms the compared algorithms on the runtime.

In magnetic resonance (MR), hardware limitation, scanning time, and patient comfort often result in the acquisition of anisotropic 3-D MR images. Enhancing image resolution is desired but has been very challenging in medical image processing. Super resolution reconstruction based on sparse representation and overcomplete dictionary has been lately employed to address this problem; however, these methods require extra training sets, which may not be always available. This paper proposes a novel single anisotropic 3-D MR image upsampling method via sparse representation and overcomplete dictionary that is trained from in-plane high resolution slices to upsample in the out-of-plane dimensions. The proposed method, therefore, does not require extra training sets. Abundant experiments, conducted on simulated and clinical brain MR images, show that the proposed method is more accurate than classical interpolation. When compared to a recent upsampling method based on the nonlocal means approach, the proposed method did not show improved results at low upsampling factors with simulated images, but generated comparable results with much better computational efficiency in clinical cases. Therefore, the proposed approach can be efficiently implemented and routinely used to upsample MR images in the out-of-planes views for radiologic assessment and postacquisition processing.

Multiple sequence alignment (MSA) is a fundamental and key step for implementing other tasks in bioinformatics, such as phylogenetic analyses, identification of conserved motifs and domains, structure prediction, etc. Despite the fact that there are many methods to implement MSA, biologically perfect alignment approaches are not found hitherto. This paper proposes a novel idea to perform MSA, where MSA is treated as a multiobjective optimization problem. A famous multiobjective evolutionary algorithm framework based on decomposition is applied for solving MSA, named MOMSA. In the MOMSA algorithm, we develop a new population initialization method and a novel mutation operator. We compare the performance of MOMSA with several alignment methods based on evolutionary algorithms, including VDGA, GAPAM, and IMSA, and also with state-of-the-art progressive alignment approaches, such as MSAprobs, Probalign, MAFFT, Procons, Clustal omega, T-Coffee, Kalign2, MUSCLE, FSA, Dialign, PRANK, and CLUSTALW. These alignment algorithms are tested on benchmark datasets BAliBASE 2.0 and BAliBASE 3.0. Experimental results show that MOMSA can obtain the significantly better alignments than VDGA, GAPAM on the most of test cases by statistical analyses, produce better alignments than IMSA in terms of TC scores, and also indicate that MOMSA is comparable with the leading progressive alignment approaches in terms of quality of alignments.

An artificial immune system inspired by the fundamental principle of the vertebrate immune system, for solving constrained optimization problems, is proposed. The analogy between the mechanism of biological immune response and constrained optimization formulation is drawn. Individuals in population are classified into feasible and infeasible groups according to their constraint violations that closely match with the two states, inactivated and activated, of B-cells in the immune response. Feasible group focuses on exploitation in the feasible areas through clonal selection, recombination, and hypermutation, while infeasible group facilitates exploration along the feasibility boundary via location update. Direction information is extracted to promote the interactions between these two groups. This approach is validated by the benchmark functions proposed most recently and compared with those of the state of the art from various branches of evolutionary computation paradigms. The performance achieved is considered fairly competitive and promising.

In this letter, we propose a novel approach of simultaneous feature and model selection for continuous hidden Markov model (CHMM). In our method, a set of real valued quantities, defined as feature saliencies, are proposed for feature selection. A variational Bayesian (VB) framework is applied to infer the feature saliencies, the number of hidden states, and the parameters of the CHMM simultaneously. Experiments based on synthetic and real data demonstrate the effectiveness of the proposed method.

Aiming to keep a balance between exploration and exploitation capability, the paper presents a particle swarm optimizer with self-adjusting neighborhoods (PSOSN). In the new algorithm, the particles are initially arranged in ring topology and then automatically adjust their own neighborhood structure based on novel neighborhood extension and restriction strategies. For efficiently controlling the process of information diffusion, neighborhood extension factor (NEF) and local impact factor (LIF) are introduced to depict particle's extension state and neighborhood relation, respectively. The experiment results demonstrate good performance of PSOSN on five benchmark functions compared with the PSO algorithms using different neighborhood schemes.

The reliability of distributed systems & computer networks in which computing nodes and/or communication links may fail with certain probabilities have been modeled by a probabilistic network. Computing the residual connectedness reliability (RCR) of probabilistic networks under the fault model with both node & link faults is very useful, but is an NP-hard problem. Up to now, there has been little research done under this fault model. There are neither accurate solutions nor heuristic algorithms for computing the RCR. In our recent research, we challenged the problem, and found efficient algorithms for the upper & lower bounds on RCR. We also demonstrated that the difference between our upper & lower bounds gradually tends to zero for large networks, and are very close to zero for small networks. These results were used in our dependable distributed system project to find a near-optimal subset of nodes to host the replicas of a critical task.

In the past few years we have developed an experimental distributed system that supports multi-task applications with different levels of criticality. Software implemented fault-tolerant protocols are used to support dependable computing. This paper first presents Markov models of a distributed system under the occurrence of faults, reconfiguration and repair. As a part of our overall project, these models are intended for solving our particular problems, like assessing the merits of redundant schemes, task allocation and reallocation policies, and fault handling used in our experimental system. However, these models are developed in a generic way. They can also be used in evaluating individual task's reliability, risk and availability under various redundant schemes in any homogeneous distributed system. Then, we extend our study in analysing the dependability of the heterogeneous system consisting of a number homogeneous distributed systems connected through gateways.

The present paper is concerned with the system-level probabilistic diagnosis problem of multiprocessor systems. First, a new diagnosis algorithm, known as the K-Step-Voting (K-SV) algorithm, is presented. This algorithm generalizes the Majority-Voting (MV) algorithm due to Blough et al. (1992). Then K-SV algorithm is theoretically proved to be better than the MV algorithm. Finally, through computer simulations, the K-SV algorithm is shown to be much superior to the MV algorithm when run on hypercube systems.