Hyperspectral image (HSI) denoising is an essential preprocessing step for downstream applications. Although Vision Transformer based approaches show impressive denoising performance through self-similarity modeling, these methods still fail to exploit spatial and spectral correlations while ensuring flexibility and efficacy. To address this issue, we propose a Hyperspectral Denoising Transformer using Low-Rank Prompt (HyLoRa), simultaneously taking the spatial self-similarity and spectral low-rank property into account for HSI denoising. Specifically, to fully utilize intrinsic similarity in spatial domain, we perform cross-shaped window based spatial self-attention for effectively modeling local and global similarity. Moreover, to exploit Low-Rank inductive bias, we integrate a Low-Rank Prompt module into attention calculation for counting corrected low-dimensional vector from a large collection of HSIs. This helps to better refine underlying noise-free structure representations. Compared to existing works, powerful capabilities for modeling spatial and spectral correlations can be built to correct low-rank representation in the feature space. Extensive experiments on both simulated and real remote sensing noise demonstrate that our HyLoRa consistently surpasses the state-of-the-art methods.

In the rapidly evolving field of consumer healthcare, the exploration of non-coding RNA interactions is crucial for drug development and personalized therapy. However, through traditional experimental validation methods, it is usually costly in terms of labor and money. In this article, we propose a efficient smart model utilizing multi-perspective weighted matrix factorization with global and local interactive-based random walk (MWMF-GLRW) to assist personalized treatment. First, only known interaction information was used to compute noncoding RNA similarities and perform fusions to ensure simplicity and generalizability of the model. Second, we innovatively develop a multi-perspective weighted matrix factorization technique. This method extracts key features while preserving the matrix structure, effectively enhancing the robustness and formative nature of miRNA-lncRNA edges between miRNA and lncRNA networks. Third, we introduce a new random walk method that considers both global information and local details of heterogeneous networks. This iterative interaction mechanism dynamically adjusts the model, enhancing its robustness and accuracy. Experiments show that MWMF-GLRW surpasses the state-of-the-art model on three datasets using only known interaction information. Notably, the simplicity of our methodology, combined with its high predictive efficiency, makes it well-suited for application in medical electronic devices aimed at promoting healthy patient consumption.

Hyperspectral images (HSIs) exhibit pronounced spatial similarity and spectral correlation. With these two physical properties taken into account, underlying clean HSI will be easier to derive from noisy images. However, existing denoising approaches struggle to model the spatial-spectral structure due to the following limitations: 1) excessive memory consumption when performing global modeling, and 2) insufficient effectiveness in local modeling. To address these issues, we propose HyMatt, a hybrid model of state space model (SSM) and attention mechanism for hyperspectral image denoising. Specifically, to fully exploit global similarity within an HSI cube, we devise Vision Mamba Quad Directions (VMamba4D) based on crafted Cube Selective Search (CSS) to capture long-range dependencies in a memory-efficient manner. Our CSS not only enhances global modeling capacity but also mitigates the negative impacts of causal modeling inherent in SSM. Furthermore, in order to improve local similarity modeling, we integrate a Local Attention (Local Attn) module, in which the adjacent elements are refined by adaptively utilizing similar neighboring features as guidance. Compared to existing methods, our HyMatt excels in exploiting local features while leveraging the global similarity within the entire HSI cube. Extensive experiments on both simulated and real remote sensing noisy images demonstrate that our HyMatt consistently surpasses the state-of-the-art HSIs denoising methods.