Due to the diversity of brain anatomy and the scarcity of annotated data, supervised anomaly detection for brain MRI remains challenging, driving the development of unsupervised anomaly detection (UAD) approaches. Current UAD methods typically utilize synthetically generated noise perturbations on healthy MRIs to train generative models for normal anatomy reconstruction, enabling anomaly detection via residual maps. However, such simulated anomalies lack the biophysical fidelity and morphological complexity characteristic of true clinical lesions. To advance UAD in brain MRI, we conduct the first systematic frequency-domain analysis of pathological signatures, revealing two key properties: (1) anomalies exhibit unique frequency patterns distinguishable from normal anatomy, and (2) low-frequency signals maintain consistent representations across healthy scans. These insights motivate our Frequency-Decomposition Preprocessing (FDP) framework—the first UAD method to leverage frequency-domain reconstruction for simultaneous pathology suppression and anatomical preservation. FDP can integrate seamlessly with existing anomaly simulation techniques, consistently enhancing detection performance across diverse architectures while maintaining diagnostic fidelity. Experimental results demonstrate that FDP consistently improves anomaly detection performance when integrated with existing methods. Notably, FDP achieves a 17.63% increase in DICE score with LDM while maintaining robust improvements across multiple baselines.

Diffusion-based virtual staining methods of histopathology images have demonstrated outstanding potential for stain normalization and cross-dye staining (e.g., hematoxylin-eosin to immunohistochemistry). However, achieving pathology-correct cross-dye virtual staining with versatile tone controls poses significant challenges due to the difficulty of decoupling the given pathology and tone conditions. This issue would cause non-pathologic regions to be mistakenly stained like pathologic ones, and vice versa, which we term “pathology leakage.” To address this issue, we propose diffusion virtual staining Transformer (D-VST), a new framework with versatile tone control for cross-dye virtual staining. Specifically, we introduce a pathology encoder in conjunction with a tone encoder, combined with a two-stage curriculum learning scheme that decouples pathology and tone conditions, to enable tone control while eliminating pathology leakage. Further, to extend our method for billion-pixel whole slide image (WSI) staining, we introduce a novel frequency-aware adaptive patch sampling strategy for high-quality yet efficient inference of ultra-high resolution images in a zero-shot manner. Integrating these two innovative components facilitates a pathology-correct, tone-controllable, cross-dye WSI virtual staining process. Extensive experiments on three virtual staining tasks that involve translating between four different dyes demonstrate the superiority of our approach in generating high-quality and pathologically accurate images compared to existing methods based on generative adversarial networks and diffusion models. Our code and trained models will be released.

As an important research topic in biomedical field, adverse drug reaction detection has received extensive research attention in the past ten years. Recent studies begin to design various neural frameworks with external knowledge to improve the task performance. However, these efforts mainly focus on extracting contextual information or knowledge according to the whole input text, which may introduce a lot of redundant information. In this paper, we propose a hybrid neural network model, which automatically integrates entity-related keywords from external knowledge base, for predicting adverse drug reaction. Experimental results on the publicly available dataset show our proposed model can achieve 93.71% and 88.07% F1 score on entity recognition and relationship extraction, respectively, outperforming the existing methods and strong neural baselines by a large margin.

Genomic variant detection is essential for cancer research and treatment, enabling precise diagnosis and personalized therapies. Considering the scarcity of labeled data, recent studies adopt few-shot learning for the task of genomic variant detection. However, the existing methods face significant challenges, including sensitivity to label distribution variability and difficulties in predicting long-span entities. To address two issues, we propose an enhanced few-shot learning method that integrates adaptive memory module, supervised contrastive learning, and orthogonal transfer module for genomic variant detection. Experimental results on the tmVar dataset demonstrate that our proposed model outperforms the existing methods and strong baselines significantly.