Medical phrase grounding is crucial for identifying relevant regions in medical images based on phrase queries, facilitating accurate image analysis and diagnosis. However, current methods rely on manual extraction of key phrases from medical reports, reducing efficiency and increasing the workload for clinicians. Additionally, the lack of model confidence estimation limits clinical trust and usability. In this paper, we introduce a novel task-Medical Report Grounding (MRG)-which aims to directly identify diagnostic phrases and their corresponding grounding boxes from medical reports in an end-to-end manner. To address this challenge, we propose uMedGround, a a robust and reliable framework that leverages a multimodal large language model to predict diagnostic phrases by embedding a unique token, < $\mathtt {BOX}$BOX >, into the vocabulary to enhance detection capabilities. A vision encoder-decoder processes the embedded token and input image to generate grounding boxes. Critically, uMedGround incorporates an uncertainty-aware prediction model, significantly improving the robustness and reliability of grounding predictions. Experimental results demonstrate that uMedGround outperforms state-of-the-art medical phrase grounding methods and fine-tuned large visual-language models, validating its effectiveness and reliability. This study represents a pioneering exploration of the MRG task, marking the first-ever endeavor in this domain. Additionally, we demonstrate the applicability of uMedGround in medical visual question answering and class-based localization tasks, where it highlights visual evidence aligned with key diagnostic phrases, supporting clinicians in interpreting various types of textual inputs, including free-text reports, visual question answering queries, and class labels.

Medical image segmentation is critical for disease diagnosis and treatment assessment. However, concerns regarding the reliability of segmentation regions persist among clinicians, mainly attributed to the absence of confidence assessment, robustness, and calibration to accuracy. To address this, we introduce deep evidential segmentation model (DEviS), an easily implementable foundational model that seamlessly integrates into various medical image segmentation networks. DEviS not only enhances the calibration and robustness of baseline segmentation accuracy but also provides high-efficiency uncertainty estimation for reliable predictions. By leveraging subjective logic theory, we explicitly model probability and uncertainty for medical image segmentation. Here, the Dirichlet distribution parameterizes the distribution of probabilities for different classes of the segmentation results. To generate calibrated predictions and uncertainty, we develop a trainable calibrated uncertainty penalty. Furthermore, DEviS incorporates an uncertainty-aware filtering (UAF) module, which designs the metric of uncertainty-calibrated error to filter out-of-distribution (OOD) data. We conducted validation studies on publicly available datasets, including ISIC2018, KiTS2021, LiTS2017, and BraTS2019, to assess the accuracy and robustness of different backbone segmentation models enhanced by DEviS, as well as the efficiency and reliability of uncertainty estimation. Additionally, two potential clinical trials were conducted using the UAF module. The clinical application conducted on the Johns Hopkins OCT and Duke OCT-DME datasets demonstrated the effectiveness of the model in filtering OOD data. The second trial evaluated its efficacy in filtering high-quality data on the FIVES datasets. At last, the proposed DEviS method was extended to semi-supervised medical image segmentation, where it exhibited strong robustness under noisy conditions. Our code has been released in https://github.com/Cocofeat/DEviS.

Dictionary learning (DL) is a pivotal task in machine learning and signal processing, involving extracting representative features from a given dataset. However, conventional DL models are known to be highly sensitive to outliers. To circumvent this issue, we introduce a new and robust DL model based on unbalanced optimal transport (UOT). Compared to DL models based on conventional robust distances and the Wasserstein distance, our model not only captures and leverages the structural information within the data but also demonstrates strong resilience to outliers. By employing the structure of the proposed robust DL model, we develop a novel hybrid block coordinate descent (BCD) algorithm. The proposed algorithm maintains computational tractability by exploiting special block structures of the subproblems. In addition, we establish the convergence of our algorithm without the Lipschitz smooth condition. Through extensive experimentation, we validate our theoretical results and demonstrate the effectiveness of the proposed method on synthetic data, MNIST data, Olivetti faces dataset, and hyperspectral images (HSIs) datasets.

The previous advancements in pathology image understanding primarily involved developing models tailored to specific tasks. Recent studies has demonstrated that the large vision-language model can enhance the performance of various downstream tasks in medical image understanding. In this study, we developed a domain-specific large language-vision assistant (PA-LLaVA) for pathology image understanding. Specifically, (1) we first construct a human pathology image-text dataset by cleaning the public medical image-text data for domain-specific alignment; (2) Using the proposed image-text data, we first train a pathology language-image pretraining (PLIP) model as the specialized visual encoder for pathology image, and then we developed scale-invariant connector to avoid the information loss caused by image scaling; (3) We adopt two-stage learning to train PA-LLaVA, first stage for domain alignment, and second stage for end to end visual question & answering (VQA) task. In experiments, we evaluate our PA-LLaVA on both supervised and zero-shot VQA datasets, our model achieved the best overall performance among multimodal models of similar scale. The ablation experiments also confirmed the effectiveness of our design. We posit that our PA-LLaVA model and the datasets presented in this work can promote research in field of computational pathology. All codes are available at: https://github.com/ddw2AIGROUP2CQUPT/PA-LLaVA

This article considers the Wasserstein barycentric coordinates problem for Gaussian distributions which is the inverse problem of the Wasserstein barycenter problem. These coordinates take into account the underlying geometry of the measure space of Gaussian distributions and are thus meaningful for applications such as diffusion analysis and distributed information fusion. When the probability supports are discrete and identical, the theory of Wasserstein barycentric coordinates is well developed. However, for general probability distributions, the computation of Wasserstein barycentric coordinates is intractable since the technical hurdles involve solving a non-convex and non-concave optimization problem. For Gaussian distributions, we derive the closed-form expression of the derivatives for the objective function and propose a projected gradient descent method to solve the problem. Finally, we illustrate its application in diffusion tensor image (DTI) denoising including simulated DTI with different noise levels and DTI of the human brain.

This paper addresses the problem of drop-point constrained state modeling in the set-valued framework, where only drop-point information of state trajectories is available. Due to the lack of complete trajectory information, existing estimation methods with linear equality constraints struggle to achieve effective state prediction. This paper primarily investigates the integration of drop-point constraint information into the entire system state evolution process via convex optimization projection in the set-valued framework, aiming to reconstruct a linear dynamic system model. Subsequently, by employing multiple affine transformations of ellipsoids and designing a weight matrix, the smoothness of state trajectories is enhanced to better align with real motion patterns. Finally, through simulation experiments, we validate the significant advantages of the reconstructed system model over traditional unconstrained system models under the set-valued framework. Additionally, we demonstrate the impact of drop-point constraints on state trajectory evolution under different initial point conditions with the same drop point.

This paper addresses the localization challenge in cooperative multi-agent wireless sensor networks, specifically focusing on range-based localization. To enhance robustness against outliers in range measurements, we employ the Huber function, leading to the formulation of a robust yet nonconvex optimization problem with coupled agent variables. Confronted with this nonconvex optimization challenge, particularly in largescale networks, we reformulate the problem using Lagrange duality and conjugate theory. This restructuring yields subproblems characterized by smooth strong convexity for dual variables and a simplified form for primal variables, thereby facilitating an efficient solution. Building upon this reformulation, we introduce a novel distributed primal-dual algorithm that employs coordinate descent and proximal minimization techniques within an iterative framework. This approach furnishes closed-form solutions for both primal and dual variables. Theoretically, our method ensures not only the convergence of the sequence of objective function values but also, by leveraging the KurdykaŁojasiewicz property, we establish the guaranteed global convergence of the location estimates sequence to a critical point of the original objective function. Notably, our proposed approach exhibits lower computational complexity, communication cost, and storage space compared to existing methods. Numerical experiments underscore the superiority of the proposed method in terms of robustness and localization accuracy when compared to the other methods in the literature.

Network topology inference is a fundamental problem in many applications of network science, such as locating the source of fake news and brain connectivity network detection. Many real-world situations suffer from a critical problem in which only a limited number of observed nodes are available. In this work, the problem of network topology inference under the framework of partial observability is considered. Based on the vector autoregression model, we propose a novel unbiased estimator for symmetric network topology with Gaussian noise and the Laplacian combination rule. Theoretically, we prove that this estimator converges in probability to the network combination matrix. Furthermore, by utilizing the Gaussian mixture model algorithm, an effective algorithm called the network inference Gauss algorithm is developed to infer the network structure. Finally, compared with state-of-the-art methods, numerical experiments demonstrate that better performance is obtained in the case of small sample sizes when using the proposed algorithm.

The objective of this study is to validate the Aqua-Moderate Resolution Imaging Spectroradiometer (MODIS) Dark Target (DT) operational aerosol products at 3 km (DT3K) and 10 km (DT10K) spatial resolutions over the South China Sea. For validation of the DT3K and DT10K AOD retrievals, the ground-based Microtops II Sun photometer AOD measurements, obtained from the Marine Aerosol Network (MAN) which were collected from 10 different cruises over the South China Sea, were used. Results showed that 38% of the DT3K and 68% of the DT10K AOD retrievals were within the expected error (EE = {+(0.04+0.10×AODMAN), -(0.02+0.10× AODMAN) } with root mean square error (RMSE) of 0.128 and 0.141 and mean absolute error (MAE) of 0.084 and 0.094, respectively. Underestimation in the DT AOD retrievals were found at both the resolutions, and a negative relationship was also found between DT AOD and normalized water reflectance. These results indicated that regional meteorology errors of the surface reflectance may have a significant effect on the underestimation of the MODIS AOD retrievals.

The distributed detection fusion is investigated for conditionally dependent sensor networks with channel errors. When the joint probability density functions of the sensor observations are dependent and high dimensional, it is known to be a challenging problem. This paper deals with this problem under Monte Carlo framework. The Bayesian cost function is approximated by Monte Carlo importance sampling. Necessary conditions for optimal sensor rules and optimal fusion rule are derived in the sense of minimizing the approximated Bayesian cost function, respectively. A Gauss-Seidel/person-by-person optimization algorithm is developed to search the optimal sensor rules. It is proved that the discretized algorithm is finitely convergent. Since the error rate of Monte Carlo integration is regardless of dimensionality, the complexity of the new algorithm is much less than that of the previous algorithm based on Riemann sum approximation. The proposed method allows us to design the sensor networks with a higher dimensional joint probability density function of the sensor observations. The typical examples with dependent observations and channel errors are examined. The results of numerical examples demonstrate the effectiveness of the new algorithm.

Multiple-source localization problem based on acoustic energy measurements is investigated by set-membership estimation theory. When the probability density function of measurement noise is unknown-but-bounded, multiple-source localization is a difficult problem since not only the acoustic energy measurement is a complicated nonlinear function of multiple sources, but also the multiple sources bring about a high-dimensional state estimation problem. The main contribution of this paper is as follows. Firstly, to deal with the nonlinear function, it is linearized by the first-order Taylor expansion with a remainder error. The point is that the bounding box of the remainder is derived in each iteration based on the convex bounding set of the state. Especially, when the state can be bounded in a cylinder, the remainder bound can be achieved analytically. Secondly, based on the separate property of the nonlinear observation function, an efficient estimation procedure is developed to deal with the high-dimensional state estimation problem by using an alternately optimization iterative algorithm. In the process of iteration, the remainder bound requires to be known on-line. Finally, a typical numerical example in multiple-source localization demonstrates the effectiveness of the set-membership localization algorithms. In particular, it shows that when the noise is non-Gaussian, the set-membership localization algorithm performs better than the maximum likelihood localization algorithm.

The set-membership information fusion problem is investigated for general multisensor nonlinear dynamic systems. Compared with linear dynamic systems and point estimation fusion in mean squared error sense, it is a more challenging nonconvex optimization problem. Usually, to solve this problem, people try to find an efficient or heuristic fusion algorithm. It is no doubt that an analytical fusion formula should be much significant for raising accuracy and reducing computational burden. However, since it is a more complicated than the convex quadratic optimization problem for linear point estimation fusion, it is not easy to get the analytical fusion formula. In order to overcome the difficulty of this problem, two popular fusion architectures are considered: centralized and distributed set-membership information fusion. Firstly, both of them can be converted into a semidefinite programming problem which can be efficiently computed, respectively. Secondly, their analytical solutions can be derived surprisingly by using decoupling technique. It is very interesting that they are quite similar in form to the classic information filter. In the two analytical fusion formulae, the information of each sensor can be clearly characterized, and the knowledge of the correlation among measurement noises across sensors are not required. Finally, multi-algorithm fusion is used to minimize the size of the state bounding ellipsoid by complementary advantages of multiple parallel algorithms. A typical numerical example in target tracking demonstrates the effectiveness of the centralized, distributed, and multi-algorithm set-membership fusion algorithms. In particular, it shows that multi-algorithm fusion performs better than the centralized and distributed fusion.

The estimation fusion problem and posterior Cramer-Rao bound (PCRB) are presented for multi-sensor nonlinear systems with uncertain observations. In order to effectively deal with the difficulties caused by uncertainty, a novel method is proposed by introducing 0-1 latent variables. It has two nice properties. Firstly, the derived estimation fusion method can take full advantage of the character of the nonlinear function and uncertain observations. Secondly, the uncertain system with a discrete variable can be approximated by a continuous system, where the discrete distribution of the latent variable is approximated by a continuous one, then the PCRB can be achieved by a limiting process of PCRB for the continuous system. Since the derived PCRB has an analytical expression, it can reduce the computational burden much more. A typical numerical example in target tracking demonstrates the effectiveness of the estimation fusion method and the proposed PCRB for the uncertain observation systems.

A novel random data association algorithm is proposed in the framework of multiple hypothesis tracking which can be equivalent to an NP-hard multidimensional assignment problem. The key idea of this new algorithm is to relax the multidimensional assignment problem to a linear programming problem. The solution of the linear programming problem may be treated as the probability of potential tracks, which avoids the computation difficulties of rounding the probability solution to 0/1 satisfying the constraints. Then all tracks and measurements can be integrated to a new whole dynamic system with random coefficient matrices. Moreover, the random coefficient matrix Kalman filtering is applied to the integrated dynamic system to derive the state estimates of the tracks. The significant advantage of the random Kalman filter-based multiple hypothesis tracking data association is that the computation complexity is much less than that of Lagrangian relaxation of the multidimensional assignment problem. Simulation demonstrates the random data association algorithm works well and the running time can be shortened greatly compared with the Lagrangian relaxation algorithm of the multidimensional assignment problem.

We consider a collaborative estimation problem using dependent observations in a wireless sensor network, where each sensor aims to maximize its estimation performance in terms of Fisher information (FI) by forming coalitions with other sensors and collaborating within a coalition. The energy consumed by the sensors increases with the size of the coalition and hence we prove that grand coalition will not form. We investigate the formation of non-overlapping coalitions such that each sensor's performance is maximized under a specific energy constraint. We decouple marginal and dependent components of FI obtained from the joint distribution by using copula theory. We introduce the concept of diversity gain and redundancy loss and demonstrate how a copula based formulation allows us to characterize these concepts. Distributed estimation problem is formulated as a coalitional game. A merge-and-split algorithm is used for finding an optimal partition. Stability of the proposed algorithm for this game is discussed. Finally, numerical results are discussed.

In this paper, a multisensor linear dynamic system with model uncertainty and bounded noises is considered. Based on previously developed set-valued estimation methods in terms of convex optimization, we propose several efficient algorithms of centralized sensor fusion, distributed sensor fusion, and multi-algorithm fusion to minimize the Euclidian estimation error of the state vector. Obviously, an ellipsoid/box with a larger “size” cannot be in general guaranteed to contain another ellipsoid/box with a smaller “size” since centers and shapes of the two ellipsoids/boxes may be different from each other. This fact and the complementary advantages of multiple sensors and multiple algorithms motivate us to construct multiple estimation ellipsoids/boxes squashed along each entry of the state vector as much as possible respectively by using the technique of multiple differently weighted objectives. Then intersection fusion of these estimation ellipsoids/boxes yields a final Euclidian-error-minimized state estimate. Numerical examples show that the new method with multi-algorithm at both the sensors and the fusion center can significantly reduce the Euclidian estimation error of the state.

In this paper, the communication direction problem of a two-sensor tandem binary decision system is considered. Rigorous analysis shows that the performance of communication from the sensor with higher noise power to the sensor with lower noise power is not always better than the performance of the reverse communication direction when the signal and sensor noises are both Gaussian. This result can be extended to a more general two-sensor tandem binary decision system without the assumption of a specific data distribution. This seems somewhat counterintuitive but has significance for optimization design of sensor communication direction. Computer experiments support our analytic results and illustrate interesting information which requires need further study.