Rare diseases pose significant diagnostic challenges due to their low prevalence, limited clinical awareness, and pronounced phenotypic heterogeneity. Early and accurate diagnosis is essential but remains difficult, especially in resource-limited settings where comprehensive genetic testing is unavailable. Distinctive facial phenotypes can offer accessible diagnostic clues, yet overlapping features and broad phenotypic spectra often hinder precise identification. To address these challenges, we present the Facial Phenotype-Gene-Disease Knowledge Graph (FPGDKG), a unified resource integrating multi-source data on facial phenotypes, genes, and diseases. The knowledge graph comprises 23,096 nodes and 239,236 relationships. We demonstrate the utility of FPGDKG through three representative use cases: (1) phenotype-based automated diagnosis of rare diseases using machine learning models; (2) explainable diagnosis by jointly presenting phenotype, genotype, and literature evidence for each prediction. The accuracies of the presented evidence, as validated quantitatively, are 73.67$\%$ for phenotype, 59.57$\%$ for gene, and 90.59$\%$ for literature evidence; (3) embedding-based matching to support differential diagnosis for ultra-rare diseases. To facilitate clinical use and research, we also developed an interactive online platform that offers intuitive visualization, information retrieval, and explainable decision support (http://bioinf.org.cn:8060/). Through three representative use cases, we show that FPGDKG supports promising diagnostic performance and enhances explainability by providing multi-dimensional evidence, making it a valuable tool for transparent, data-driven rare disease diagnosis.

Accurate and actionable prognostic models can meaningfully influence follow-up scheduling, therapeu tic prioritization, and resource allocation in oncology. We propose GLLM, a multimodal graph learning framework that integrates RNA-seq profiles, routine clinical variables, and structural protein embeddings derived from protein language models to stratify patients by 5-year risk across multiple cancer types. Each gene is represented as a node within a protein-protein interaction graph, and we intro duce SCANE, a fusion mechanism that modulates each gene's structural embedding using patient-specific expression values. This design enables the graph neural network to propagate expression-conditioned molecular sig nals while preserving the underlying biophysical context. Across breast cancer, lung adenocarcinoma, and colorectal cancer cohorts, GLLM improves the area under the precision-recall curve relative to strong clinical and molecular baselines, while maintaining competitive concordance indices. The contributions of this work include: (1) an effective fusion strategy that enhances node representations by combining protein structural embeddings with gene expression for improved risk prediction; (2) a sys tematic evaluation demonstrating that sequence-derived structural embeddings outperform text-based biomedical embeddings; and (3) patient-level interpretability analyses showing that the model highlights established biomarkers and aligns with perturbation-based sensitivity profiles. Clinical significance: GLLM supports personalized surveillance planning by identifying high-risk patients who may benefit from earlier imaging, shorter follow-up inter vals, or prioritization for treatment discussions and clinical trial screening. Its lightweight architecture (<7 MFLOPs) enables seamless integration into existing oncology work f lows without additional computational burden. The result ing risk score is designed to complement, rather than replace, mutation profiling and clinicopathological staging, reflecting the biological and operational heterogeneity across cancer types.

Accurate segmentation of medical images, particularly for anatomical structures with irregular shapes and low contrast such as the esophagus, remains a significant challenge. To address this issue, we propose MEM-UNet, a robust 3D Mamba-based UNet framework enhanced by mathematical morphology. Our approach adapts the State Space Model (SSM) in Mamba to support three-dimensional CT volumes, establishing an effective 3D perception backbone for the UNet architecture. In addition, we incorporate Morphology-Aware Spatial-Channel Attention (MASCA) blocks into the skip connections, where Morphology-Enhanced Spatial Convolution (MESC) augments spatial representations while Squeeze-and-Excitation (SE) highlights channel- wise features. This integration effectively leverages the shape-awareness provided by morphological operations, thus improving boundary precision. To further refine segmentation, we introduce a Morphology-Enhanced Decision (MED) layer that sharpens contour boundaries and performs voxel-level classification with high precision. Extensive experiments on SegTHOR and BTCV datasets demonstrate that MEM-UNet surpasses state-of-the-art models, achieving Dice Similarity Coefficient (DSC) scores of 87.42% and 74.86% for multi-organ segmentation, and 78.94% and 67.70% for esophagus segmentation, respectively. Ablation studies confirm the effectiveness of the proposed components and highlight the benefits of integrating mathematical morphology into our pipeline. The implementation is available at https://gitfront.io/r/cheee123/DDTJhrf3LRMd/MEM-UNet/.

The interaction between mothers and young children is a highly dynamic process neurally characterized by inter-brain synchrony (IBS) at θ and/or α rhythms. However, their establishment, dynamic changes, and roles in mother-child interactions remain unknown. In this study, through a simultaneous dynamic analysis of inter-brain EEG synchrony, intra-brain EEG power, and interactive behaviors from 40 mother-preschooler dyads during turn-taking cooperation, we constructed a dynamic inter-brain model that θ-IBS and α-IBS alternated with interactive behaviors, with EEG frequency-shift as a prerequisite for IBS transitions. When mothers attempt to track their children's attention and/or predict their intentions, they will adjust their EEG frequencies to align with their children's θ oscillations, leading to a higher occurrence of the θ-IBS state. Conversely, the α-IBS state, accompanied by the EEG frequency-shift to the α range, is more prominent during mother-led interactions. Further exploratory analysis reveals greater presence and stability of the θ-IBS state during cooperative than non-cooperative conditions, particularly in dyads with stronger emotional attachments and more frequent interactions in their daily lives. Our findings shed light on the neural oscillatory substrates underlying the IBS dynamics during mother-preschooler interactions.

As the impact of chronic mental disorders increases, multimodal sentiment analysis (MSA) has emerged to improve diagnosis and treatment. In this paper, our approach leverages disentangled representation learning to address modality heterogeneity with self-supervised learning as a guidance. The self-supervised learning is proposed to generate pseudo unimodal labels and guide modality-specific representation learning, preventing the acquisition of meaningless features. Additionally, we also propose a text-centric fusion to effectively mitigate the impacts of noise and redundant information and fuse the acquired disentangled representations into a comprehensive multimodal representation. We evaluate our model on three publicly available benchmark datasets for multimodal sentiment analysis and a privately collected dataset focusing on schizophrenia counseling. The experimental results demonstrate state-of-the-art performance across various metrics on the benchmark datasets, surpassing related works. Furthermore, our learning algorithm shows promising performance in real-world applications, outperforming our previous work and achieving significant progress in schizophrenia assessment.

Low reliability has consistently been a challenge in the application of deep learning models for high-risk decision-making scenarios. In medical image segmentation, multiple expert annotations can be consulted to reduce subjective bias and reach a consensus, thereby enhancing the segmentation accuracy and reliability. To develop a reliable lesion segmentation model, we propose CalDiff, a novel framework that can leverage the uncertainty from multiple annotations, capture real-world diagnostic variability and provide more informative predictions. To harness the superior generative ability of diffusion models, a dual step-wise and sequence-aware calibration mechanism is proposed on the basis of the sequential nature of diffusion models. We evaluate the calibrated model through a comprehensive quantitative and visual analysis, addressing the previously overlooked challenge of assessing uncertainty calibration and model reliability in scenarios with multiple annotations and multiple predictions. Experimental results on two lesion segmentation datasets demonstrate that CalDiff produces uncertainty maps that can reflect low confidence areas, further indicating the false predictions made by the model. By calibrating the uncertainty in the training phase, the uncertain areas produced by our model are closely correlated with areas where the model has made errors in the inference. In summary, the uncertainty captured by CalDiff can serve as a powerful indicator, which can help mitigate the risks of adopting model's outputs, allowing clinicians to prioritize reviewing areas or slices with higher uncertainty and enhancing the model's reliability and trustworthiness in clinical practice.

Intracranial aneurysms (ICA) commonly occur in specific segments of the Circle of Willis (CoW), primarily, onto thirteen major arterial bifurcations. An accurate detection of these critical landmarks is necessary for a prompt and efficient diagnosis. We introduce a fully automated landmark detection approach for CoW bifurcations using a two-step neural networks process. Initially, an object detection network identifies regions of interest (ROIs) proximal to the landmark locations. Subsequently, a modified U-Net with deep supervision is exploited to accurately locate the bifurcations. This two-step method reduces various problems, such as the missed detections caused by two landmarks being close to each other and having similar visual characteristics, especially when processing the complete MRA Time-of-Flight (TOF). Additionally, it accounts for the anatomical variability of the CoW, which affects the number of detectable landmarks per scan. We assessed the effectiveness of our approach using two cerebral MRA datasets: our In-House dataset which had varying numbers of landmarks, and a public dataset with standardized landmark configuration. Our experimental results demonstrate that our method achieves the highest level of performance on a bifurcation detection task.

Multimodal magnetic resonance imaging (MRI) is instrumental in differentiating liver lesions. The major challenge involves modeling reliable connections and simultaneously learning complementary information across various MRI sequences. While previous studies have primarily focused on multimodal integration in a pair-wise manner using few modalities, our research seeks to advance a more comprehensive understanding of interaction modeling by establishing complex high-order correlations among the diverse modalities in multimodal MRI. In this paper, we introduce a multimodal graph learning with multi-hypergraph reasoning network to capture the full spectrum of both pair-wise and group-wise relationships among different modalities. Specifically, a weight-shared encoder extracts features from regions of interest (ROI) images across all modalities. Subsequently, a collection of uniform hypergraphs are constructed with varying vertex configurations, allowing for the modeling of not only pair-wise correlations but also the high-order collaborations for relational reasoning. Following information propagation through the hypergraph message passing, adaptive intra-modality fusion module is proposed to effectively fuse feature representations from different hypergraphs of the same modality. Finally, all refined features are concatenated to prepare for the classification task. Our experimental evaluations, including focal liver lesions classification using the LLD-MMRI2023 dataset and early recurrence prediction of hepatocellular carcinoma using our internal datasets, demonstrate that our method significantly surpasses the performance of existing approaches, indicating the effectiveness of our model in handling both pair-wise and group-wise interactions across multiple modalities.

Medical imaging has developed from an auxiliary means of clinical examination into a significant method and intuitive basis for clinical diagnosis of diseases, providing all-around and full-cycle health protection for the people. The Internet of Medical Things (IoMT) allows medical equipment, intelligent terminals, medical infrastructure, and other elements of medical production to be interconnected, eliminating information silos and data fragmentation. Medical images disseminated in IoMT contain a wide diversity of sensitive patient information, which means protecting the patient's personal information is vital. In this work, an Adversarial-improved reversible steganography network (Airs-Net) for computed tomography (CT) images in the IoMT is presented. Specifically, the Airs-Net adopting the prediction-embedding strategy mainly consists of an image restoration network, an embedded pixel location network, and a discriminator. The image restoration network is effective in restoring the pixel prediction error of the restoration set in integer and non-integer scaled images of arbitrary size when information is concealed. The embedded information location network can automatically select pixel locations for information embedding based on the interpolated image features of the degraded image. The restored image, embedding location map, and embedding information are fed into the embedder for information embedding, and the subsequent secret-carrying image is continuously optimized for the quality of the information-embedded image by the discriminator. Quantitative results show that Airs-Net outperforms state-of-the-art methods in both PSNR and SSIM. Further, the qualitative and quantitative results and analyses under specific clinical application scenarios and in coping with multiple types of medical image information hiding demonstrate the excellent generalization performance and practical application capability of the Airs-Net.

Uncertainty-aware consistency learning is one of the reliable approaches in semi-supervised medical image segmentation, enforcing robust model predictions under various perturbations. However, existing methods often rely on multiple stochastic predictions or dual-network/decoder discrepancies to estimate uncertainty, which increases computational cost and discards uncertain regions, potentially missing complex structures such as ambiguous lesion boundaries. To address these challenges, we introduce a Dual Uncertainty-Guided Consistency and Regional Contrastive Learning (DUCore) framework. DUCore improves segmentation robustness by integrating two complementary loss functions within consistency learning. The dual uncertainty-guided consistency loss (DuCL) adaptively calibrates the prediction alignment by prioritizing uncertain regions. DuCL uses deterministic single-pass uncertainty estimation, employing entropy-based calibration for aleatoric uncertainty and Proxy Dirichlet calibration for epistemic uncertainty. These uncertainty measures are computed directly from network output, and moderately uncertain regions are weighted instead of being discarded, which preserves valuable learning signals. The Regional Contrastive Loss (ReCL) further refines feature separability using boundary- and gradient-based hard negative mining in the encoded representation space. By explicitly targeting structural ambiguities, ReCL distinguishes lesion and organ edges from visually similar boundary-adjacent regions and mitigates intensity overlaps in gradient-rich transitions. As a result, DUCore is able to delineate fine structures and complex boundaries with higher precision. Extensive experiments on various medical segmentation benchmarks reveal that DUCore outperforms existing consistency methods.