Medical image segmentation provides useful information about the shape and size of organs, which is beneficial for improving diagnosis, analysis, and treatment. Despite traditional deep learning-based models can extract domain-specific knowledge, they face a generalization bottleneck due to the limited embedded knowledge scope. Vision foundation models have been demonstrated to be effective in extracting generalizable knowledge, but they cannot extract domain-specific knowledge without fine-tuning. In this work, we propose a novel multi-view evidential learning-based framework, which can extract both domain-specific and generalizable knowledge from multi-view features by combining the advantages of traditional and vision foundation models. Specifically, a novel multi-view state space model (MV-SSM) is designed to extract task-related knowledge while removing redundant information within multi-view features. The proposed MV-SSM utilizes Mamba, a state space model, to model cross-view contextual dependencies between domain-specific and generalizable features. Additionally, evidential learning is adopted to quantify the segmentation uncertainty of the model for boundary. In special, variational Dirichlet is introduced to characterize the distribution of the result probabilities, parameterized with collected evidence to quantify uncertainty. As a result, the model can reduce the segmentation uncertainties of boundaries by optimizing the parameters of the Dirichlet distribution. Experimental results on three datasets show that our method obtains superior segmentation performance.

With advancements in visual language models (VLMs) and large language models (LLMs), video anomaly detection (VAD) has progressed beyond binary classification to fine-grained categorization and multidimensional analysis. However, existing methods focus mainly on coarse-grained detection, lacking anomaly explanations. To address these challenges, we propose Ex-VAD, an Explainable Fine-grained Video Anomaly Detection approach that combines fine-grained classification with detailed explanations of anomalies. First, we use a VLM to extract frame-level captions, and an LLM converts them to video-level explanations, enhancing the model's explainability. Second, integrating textual explanations of anomalies with visual information greatly enhances the model's anomaly detection capability. Finally, we apply label-enhanced alignment to optimize feature fusion, enabling precise fine-grained detection. Extensive experimental results on the UCF-Crime and XD-Violence datasets demonstrate that Ex-VAD significantly outperforms existing State-of-The-Art methods.

In microbiome analysis, researchers often seek to identify taxonomic features associated with an outcome of interest. However, microbiome features are intercorrelated and linked by phylogenetic relationships, making it challenging to assess the association between an individual feature and an outcome. This paper proposes a novel conditional association test, CAT, that can account for other features and phylogenetic relatedness when testing the association between a feature and an outcome. CAT adopts a permutation approach, measuring the importance of a feature in predicting the outcome by permuting operational taxonomic unit/amplicon sequence variant counts belonging to that feature from the data and quantifying how much the association with the outcome is weakened through the change in the coefficient of determination $R^{2}$. Compared with marginal association tests, it focuses on the added value of a feature in explaining outcome variation that is not captured by other features. By leveraging global tests including PERMANOVA and MiRKAT-based methods, CAT allows association testing for continuous, binary, categorical, count, survival, and correlated outcomes. We demonstrate through simulation studies that CAT can provide a direct quantification of feature importance that is distinct from that of marginal association tests, and illustrate CAT with applications to two real-world studies on the microbiome in melanoma patients: one examining the role of the microbiome in shaping immunotherapy response, and one investigating the association between the microbiome and survival outcomes. Our results illustrate the potential of CAT to inform the design of microbiome interventions aimed at improving clinical outcomes.

BACKGROUND: Identification of features is a critical task in microbiome studies that is complicated by the fact that microbial data are high dimensional and heterogeneous. Masked by the complexity of the data, the problem of separating signals (differential features between groups) from noise (features that are not differential between groups) becomes challenging and troublesome. For instance, when performing differential abundance tests, multiple testing adjustments tend to be overconservative, as the probability of a type I error (false positive) increases dramatically with the large numbers of hypotheses. Moreover, the grouping effect of interest can be obscured by heterogeneity. These factors can incorrectly lead to the conclusion that there are no differences in the microbiome compositions. RESULTS: We translate and represent the problem of identifying differential features, which are differential in two-group comparisons (e.g., treatment versus control), as a dynamic layout of separating the signal from its random background. More specifically, we progressively permute the grouping factor labels of the microbiome samples and perform multiple differential abundance tests in each scenario. We then compare the signal strength of the most differential features from the original data with their performance in permutations, and will observe a visually apparent decreasing trend if these features are true positives identified from the data. Simulations and applications on real data show that the proposed method creates a U-curve when plotting the number of significant features versus the proportion of mixing. The shape of the U-Curve can convey the strength of the overall association between the microbiome and the grouping factor. We also define a fragility index to measure the robustness of the discoveries. Finally, we recommend the identified features by comparing p-values in the observed data with p-values in the fully mixed data. CONCLUSIONS: We have developed this into a user-friendly and efficient R-shiny tool with visualizations. By default, we use the Wilcoxon rank sum test to compute the p-values, since it is a robust nonparametric test. Our proposed method can also utilize p-values obtained from other testing methods, such as DESeq. This demonstrates the potential of the progressive permutation method to be extended to new settings.

SUMMARY: In fields, such as ecology, microbiology and genomics, non-Euclidean distances are widely applied to describe pairwise dissimilarity between samples. Given these pairwise distances, principal coordinates analysis is commonly used to construct a visualization of the data. However, confounding covariates can make patterns related to the scientific question of interest difficult to observe. We provide adjusted principal coordinates analysis as an easy-to-use tool, available as both an R package and a Shiny app, to improve data visualization in this context, enabling enhanced presentation of the effects of interest. AVAILABILITY AND IMPLEMENTATION: The R package 'aPCoA' and Shiny app can be accessed at https://cran.r-project.org/web/packages/aPCoA/index.html and https://biostatistics.mdanderson.org/shinyapps/aPCoA/.