Search

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}{https://github.com/ddw2AIGROUP2CQUPT/PA-LLaVA, Comment: 8 pages, 4 figs

Multi-modal ophthalmic image classification plays a key role in diagnosing eye diseases, as it integrates information from different sources to complement their respective performances. However, recent improvements have mainly focused on accuracy, often neglecting the importance of confidence and robustness in predictions for diverse modalities. In this study, we propose a novel multi-modality evidential fusion pipeline for eye disease screening. It provides a measure of confidence for each modality and elegantly integrates the multi-modality information using a multi-distribution fusion perspective. Specifically, our method first utilizes normal inverse gamma prior distributions over pre-trained models to learn both aleatoric and epistemic uncertainty for uni-modality. Then, the normal inverse gamma distribution is analyzed as the Student's t distribution. Furthermore, within a confidence-aware fusion framework, we propose a mixture of Student's t distributions to effectively integrate different modalities, imparting the model with heavy-tailed properties and enhancing its robustness and reliability. More importantly, the confidence-aware multi-modality ranking regularization term induces the model to more reasonably rank the noisy single-modal and fused-modal confidence, leading to improved reliability and accuracy. Experimental results on both public and internal datasets demonstrate that our model excels in robustness, particularly in challenging scenarios involving Gaussian noise and modality missing conditions. Moreover, our model exhibits strong generalization capabilities to out-of-distribution data, underscoring its potential as a promising solution for multimodal eye disease screening., Comment: 27 pages, 7 figures, 9 tables

Medical Report Grounding is pivotal in identifying the most relevant regions in medical images based on a given phrase query, a critical aspect in medical image analysis and radiological diagnosis. However, prevailing visual grounding approaches necessitate the manual extraction of key phrases from medical reports, imposing substantial burdens on both system efficiency and physicians. In this paper, we introduce a novel framework, Medical Report Grounding (MedRG), an end-to-end solution for utilizing a multi-modal Large Language Model to predict key phrase by incorporating a unique token, BOX, into the vocabulary to serve as an embedding for unlocking detection capabilities. Subsequently, the vision encoder-decoder jointly decodes the hidden embedding and the input medical image, generating the corresponding grounding box. The experimental results validate the effectiveness of MedRG, surpassing the performance of the existing state-of-the-art medical phrase grounding methods. This study represents a pioneering exploration of the medical report grounding task, marking the first-ever endeavor in this domain., Comment: 12 pages, 4 figures

Directional motion towards a specified destination is a common occurrence in physical processes and human societal activities. Utilizing this prior information can significantly improve the control and predictive performance of system models. This paper primarily focuses on reconstructing linear dynamic system models based on destination constraints in the set-valued framework. We treat destination constraints as inherent information in the state evolution process and employ convex optimization techniques to construct a coherent and robust state model. This refined model effectively captures the impact of destination constraints on the state evolution at each time step. Furthermore, we design an optimal weight matrix for the reconstructed model to ensure smoother and more natural trajectories of state evolution. We also analyze the theoretical guarantee of optimality for this weight matrix and the properties of the reconstructed model. Finally, simulation experiments verify that the reconstructed model has significant advantages over the unconstrained and unoptimized weighted models and constrains the evolution of state trajectories with different starting and ending points., Comment: 15 pages, 11 figures

The nonconvex and nonsmooth finite-sum optimization problem with linear constraint has attracted much attention in the fields of artificial intelligence, computer, and mathematics, due to its wide applications in machine learning and the lack of efficient algorithms with convincing convergence theories. A popular approach to solve it is the stochastic Alternating Direction Method of Multipliers (ADMM), but most stochastic ADMM-type methods focus on convex models. In addition, the variance reduction (VR) and acceleration techniques are useful tools in the development of stochastic methods due to their simplicity and practicability in providing acceleration characteristics of various machine learning models. However, it remains unclear whether accelerated SVRG-ADMM algorithm (ASVRG-ADMM), which extends SVRG-ADMM by incorporating momentum techniques, exhibits a comparable acceleration characteristic or convergence rate in the nonconvex setting. To fill this gap, we consider a general nonconvex nonsmooth optimization problem and study the convergence of ASVRG-ADMM. By utilizing a well-defined potential energy function, we establish its sublinear convergence rate $O(1/T)$, where $T$ denotes the iteration number. Furthermore, under the additional Kurdyka-Lojasiewicz (KL) property which is less stringent than the frequently used conditions for showcasing linear convergence rates, such as strong convexity, we show that the ASVRG-ADMM sequence has a finite length and converges to a stationary solution with a linear convergence rate. Several experiments on solving the graph-guided fused lasso problem and regularized logistic regression problem validate that the proposed ASVRG-ADMM performs better than the state-of-the-art methods., Comment: 40 Pages, 8 figures

Multisensor track-to-track fusion for target tracking involves two primary operations: track association and estimation fusion. For estimation fusion, lossless measurement transformation of sensor measurements has been proposed for single target tracking. In this paper, we investigate track association which is a fundamental and important problem for multitarget tracking. First, since the optimal track association problem is a multi-dimensional assignment (MDA) problem, we demonstrate that MDA-based data association (with and without prior track information) using linear transformations of track measurements is lossless, and is equivalent to that using raw track measurements. Second, recent superior scalability and performance of belief propagation (BP) algorithms enable new real-time applications of multitarget tracking with resource-limited devices. Thus, we present a BP-based multisensor track association method with transformed measurements and show that it is equivalent to that with raw measurements. Third, considering communication constraints, it is more beneficial for local sensors to send in compressed data. Two analytical lossless transformations for track association are provided, and it is shown that their communication requirements from each sensor to the fusion center are less than those of fusion with raw track measurements. Numerical examples for tracking an unknown number of targets verify that track association with transformed track measurements has the same performance as that with raw measurements and requires fewer communication bandwidths.

Classification and segmentation are crucial in medical image analysis as they enable accurate diagnosis and disease monitoring. However, current methods often prioritize the mutual learning features and shared model parameters, while neglecting the reliability of features and performances. In this paper, we propose a novel Uncertainty-informed Mutual Learning (UML) framework for reliable and interpretable medical image analysis. Our UML introduces reliability to joint classification and segmentation tasks, leveraging mutual learning with uncertainty to improve performance. To achieve this, we first use evidential deep learning to provide image-level and pixel-wise confidences. Then, an Uncertainty Navigator Decoder is constructed for better using mutual features and generating segmentation results. Besides, an Uncertainty Instructor is proposed to screen reliable masks for classification. Overall, UML could produce confidence estimation in features and performance for each link (classification and segmentation). The experiments on the public datasets demonstrate that our UML outperforms existing methods in terms of both accuracy and robustness. Our UML has the potential to explore the development of more reliable and explainable medical image analysis models. We will release the codes for reproduction after acceptance., Comment: 13 pages

Multimodality eye disease screening is crucial in ophthalmology as it integrates information from diverse sources to complement their respective performances. However, the existing methods are weak in assessing the reliability of each unimodality, and directly fusing an unreliable modality may cause screening errors. To address this issue, we introduce a novel multimodality evidential fusion pipeline for eye disease screening, EyeMoSt, which provides a measure of confidence for unimodality and elegantly integrates the multimodality information from a multi-distribution fusion perspective. Specifically, our model estimates both local uncertainty for unimodality and global uncertainty for the fusion modality to produce reliable classification results. More importantly, the proposed mixture of Student's $t$ distributions adaptively integrates different modalities to endow the model with heavy-tailed properties, increasing robustness and reliability. Our experimental findings on both public and in-house datasets show that our model is more reliable than current methods. Additionally, EyeMost has the potential ability to serve as a data quality discriminator, enabling reliable decision-making for multimodality eye disease screening., Comment: MICCAI 2023 (Early accept):11 pages, 4 figures

The use of AI systems in healthcare for the early screening of diseases is of great clinical importance. Deep learning has shown great promise in medical imaging, but the reliability and trustworthiness of AI systems limit their deployment in real clinical scenes, where patient safety is at stake. Uncertainty estimation plays a pivotal role in producing a confidence evaluation along with the prediction of the deep model. This is particularly important in medical imaging, where the uncertainty in the model's predictions can be used to identify areas of concern or to provide additional information to the clinician. In this paper, we review the various types of uncertainty in deep learning, including aleatoric uncertainty and epistemic uncertainty. We further discuss how they can be estimated in medical imaging. More importantly, we review recent advances in deep learning models that incorporate uncertainty estimation in medical imaging. Finally, we discuss the challenges and future directions in uncertainty estimation in deep learning for medical imaging. We hope this review will ignite further interest in the community and provide researchers with an up-to-date reference regarding applications of uncertainty estimation models in medical imaging., Comment: 11 pages, 3 figures, 3 tables

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 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 the problem of 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 module, which utilizes the metric of uncertainty-calibrated error to filter reliable data within the dataset. We conducted validation studies to assess both the accuracy and robustness of DEviS segmentation, along with evaluating the efficiency and reliability of uncertainty estimation. These evaluations were performed using publicly available datasets including ISIC2018, LiTS2017, and BraTS2019. Additionally, two potential clinical trials are being conducted at Johns Hopkins OCT, Duke-OCT-DME, and FIVES datasets to demonstrate their efficacy in filtering high-quality or out-of-distribution data. Our code has been released in https://github.com/Cocofeat/DEviS., Comment: 34 pages, 11 figures

This paper considers the state estimation problem for nonlinear dynamic systems with unknown but bounded noises. Set membership filter (SMF) is a popular algorithm to solve this problem. In the set membership setting, we investigate the filter problem where the state estimation requires to be constrained by a linear or nonlinear equality. We propose a consensus alternating direction method of multipliers (ADMM) based SMF algorithm for nonlinear dynamic systems. To deal with the difficulty of nonlinearity, instead of linearizing the nonlinear system, a semi-infinite programming (SIP) approach is used to transform the nonlinear system into a linear one, which allows us to obtain a more accurate estimation ellipsoid. For the solution of the SIP, an ADMM algorithm is proposed to handle the state estimation constraints, and each iteration of the algorithm can be solved efficiently. Finally, the proposed filter is applied to typical numerical examples to demonstrate its effectiveness.

Distributed cooperative localization in wireless networks is a challenging problem since it typically requires solving a large-scale nonconvex and nonsmooth optimization problem. In this paper, we reformulate the classic cooperative localization problem as a smooth and constrained nonconvex minimization problem while its loss function is separable over nodes. By utilizing the structure of the reformulation, we propose two novel scaled proximal alternating direction method of multipliers (SP-ADMM) algorithms, which can be implemented in a distributed manner. Compared with the classic semi-definite programming relaxation techniques, the proposed algorithms can provide more accurate position estimates with significantly lower computation complexity. The associated theoretical analysis shows that our algorithms {\blue globally converge to a KKT point} of the reformulated problem and a critical point of the original problem, with a favorable sublinear $\mathcal{O}\left(1/T\right)$ convergence rate, where $T$ is the iteration counter. Numerical experiments have consistently shown that the proposed SP-ADMM algorithms are superior to state-of-the-art methods in terms of localization accuracy and computational complexity across all tested scenarios, varying network size, number of anchors, average number of neighbors, and noise variance levels.

This paper considers the problem of tracking a large-scale number of group targets. Usually, multi-target in most tracking scenarios are assumed to have independent motion and are well-separated. However, for group target tracking (GTT), the targets within groups are closely spaced and move in a coordinated manner, the groups can split or merge, and the numbers of targets in groups may be large, which lead to more challenging data association, filtering and computation problems. Within the belief propagation (BP) framework, we propose a scalable group target belief propagation (GTBP) method by jointly inferring target existence variables, group structure, data association and target states. The method can efficiently calculate the approximations of the marginal posterior distributions of these variables by performing belief propagation on the devised factor graph. As a consequence, GTBP is capable of capturing the changes in group structure, e.g., group splitting and merging. Furthermore, we model the evolution of targets as the co-action of the group or single-target motions specified by the possible group structures and corresponding probabilities. This flexible modeling enables seamless and simultaneous tracking of multiple group targets and ungrouped targets. Particularly, GTBP has excellent scalability and low computational complexity. It not only maintains the same scalability as BP, i.e., scaling linearly in the number of sensor measurements and quadratically in the number of targets, but also only scales linearly in the number of preserved group partitions. Finally, numerical experiments are presented to demonstrate the effectiveness and scalability of the proposed GTBP method.

Despite recent improvements in the accuracy of brain tumor segmentation, the results still exhibit low levels of confidence and robustness. Uncertainty estimation is one effective way to change this situation, as it provides a measure of confidence in the segmentation results. In this paper, we propose a trusted brain tumor segmentation network which can generate robust segmentation results and reliable uncertainty estimations without excessive computational burden and modification of the backbone network. In our method, uncertainty is modeled explicitly using subjective logic theory, which treats the predictions of backbone neural network as subjective opinions by parameterizing the class probabilities of the segmentation as a Dirichlet distribution. Meanwhile, the trusted segmentation framework learns the function that gathers reliable evidence from the feature leading to the final segmentation results. Overall, our unified trusted segmentation framework endows the model with reliability and robustness to out-of-distribution samples. To evaluate the effectiveness of our model in robustness and reliability, qualitative and quantitative experiments are conducted on the BraTS 2019 dataset., Comment: 11 pages, 4 figures, Accepted by MICCAI 2022

Network topology inference is a fundamental problem in many applications of network science, such as locating the source of fake news, brain connectivity networks detection, etc. Many real-world situations suffer from a critical problem that only a limited part of observed nodes are available. This letter considers the problem of network topology inference under the framework of partial observability. Based on the vector autoregressive model, we propose a novel unbiased estimator for the symmetric network topology with the Gaussian noise and the Laplacian combination rule. Theoretically, we prove that it converges to the network combination matrix in probability. Furthermore, by utilizing the Gaussian mixture model algorithm, an effective algorithm called network inference Gauss algorithm is developed to infer the network structure. Finally, compared with the state-of-the-art methods, numerical experiments demonstrate the proposed algorithm enjoys better performance in the case of small sample sizes., Comment: 9 pages, 4 figures

Differential privacy (DP) is an essential technique for privacy-preserving. It was found that a large model trained for privacy preserving performs worse than a smaller model (e.g. ResNet50 performs worse than ResNet18). To better understand this phenomenon, we study high dimensional DP learning from the viewpoint of generalization. Theoretically, we show that for the simple Gaussian model with even small DP noise, if the dimension is large enough, then the classification error can be as bad as the random guessing. Then we propose a feature selection method to reduce the size of the model, based on a new metric which trades off the classification accuracy and privacy preserving. Experiments on real data support our theoretical results and demonstrate the advantage of the proposed method.

This paper considers the data association problem for multi-target tracking. Multiple hypothesis tracking is a popular algorithm for solving this problem but it is NP-hard and is is quite complicated for a large number of targets or for tracking maneuvering targets. To improve tracking performance and enhance robustness, we propose a randomized multiple model multiple hypothesis tracking method, which has three distinctive advantages. First, it yields a randomized data association solution which maximizes the expectation of the logarithm of the posterior probability and can be solved efficiently by linear programming. Next, the state estimation performance is improved by the random coefficient matrices Kalman filter, which mitigates the difficulty introduced by randomized data association, i.e., where the coefficient matrices of the dynamic system are random. Third, the probability that the target follows a specific dynamic model is derived by jointly optimizing the multiple possible models and data association hypotheses, and it does not require prior mode transition probabilities. Thus, it is more robust for tracking multiple maneuvering targets. Simulations demonstrate the efficiency and superior results of the proposed algorithm over interacting multiple model multiple hypothesis tracking., Comment: 28 pages, 3 figures

Extreme public health interventions play a critical role in mitigating the local and global prevalence and pandemic potential of COVID-19. Here, we use population size for pathogen transmission to measure the intensity of public health interventions, which is a key characteristic variable for nowcasting and forecasting of the epidemic. By formulating a hidden Markov dynamic system and using nonlinear filtering theory, we have developed a stochastic epidemic dynamic model under public health interventions. The model parameters and states are estimated in time from internationally available public data by combining an unscented filter and an interacting multiple model filter. Moreover, we consider the computability of the population size and provide its selection criterion. We estimate the mean of the basic reproductive number of China and the rest of the globe except China (GEC) to be 2.46 (95% CI: 2.41-2.51) and 3.64 (95% CI: (3.55-3.72), respectively. We infer that the number of latent infections of GEC is about 7.47*10^5 (95% CI: 7.32*10^5-7.62*10^5) as of April 2, 2020. We predict that the peak of infections in hospitals of GEC may reach 3.00*10^6 on the present trajectory, i.e., if the population size for pathogen transmission and epidemic parameters remains unchanged. If the control intensity is strengthened, e.g., 50% reduction or 75% reduction of the population size for pathogen transmission, the peak would decline to 1.84*10^6, 1.27*10^6, respectively., Comment: 30 pages, 10 figures

In this paper, the multiple-source ellipsoidal localization problem based on acoustic energy measurements is investigated via 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 measurements are complicated nonlinear functions of multiple sources, but also the multiple sources bring about a high-dimensional state estimation problem. First, when the energy parameter and the position of the source are bounded in an interval and a ball respectively, the nonlinear remainder bound of the Taylor series expansion is obtained analytically on-line. Next, based on the separability of the nonlinear measurement function, an efficient estimation procedure is developed. It solves the multiple-source localization problem by using an alternating optimization iterative algorithm, in which the remainder bound needs to be known on-line. For this reason, we first derive the remainder bound analytically. When the energy decay factor is unknown but bounded, an efficient estimation procedure is developed based on interval mathematics. Finally, numerical examples demonstrate the effectiveness of the ellipsoidal localization algorithms for multiple-source localization. In particular, our results show that when the noise is non-Gaussian, the set-membership localization algorithm performs better than the EM localization algorithm., Comment: 20 pages, 13 figures, submitted to Automatica

In this paper, we propose a dual set membership filter for nonlinear dynamic systems with unknown but bounded noises, and it has three distinctive properties. Firstly, the nonlinear system is translated into the linear system by leveraging a semi-infinite programming, rather than linearizing the nonlinear function. In fact, the semi-infinite programming is to find an ellipsoid bounding the nonlinear transformation of an ellipsoid, which aims to compute a tight ellipsoid to cover the state. Secondly, the duality result of the semi-infinite programming is derived by a rigorous analysis, then a first order Frank-Wolfe method is developed to efficiently solve it with a lower computation complexity. Thirdly, the proposed filter can take advantage of the linear set membership filter framework and can work on-line without solving the semidefinite programming problem. Furthermore, we apply the dual set membership filter to a typical scenario of mobile robot localization. Finally, two illustrative examples in the simulations show the advantages and effectiveness of the dual set membership filter., Comment: 26 pages, 9 figures