Histopathological whole slide images (WSIs) classification has become a foundation task in medical microscopic imaging processing. Prevailing approaches involve learning WSIs as instance-bag representations, emphasizing significant instances but struggling to capture the interactions between instances. Additionally, conventional graph representation methods utilize explicit spatial positions to construct topological structures but restrict the flexible interaction capabilities between instances at arbitrary locations, particularly when spatially distant. In response, we propose a novel dynamic graph representation algorithm that conceptualizes WSIs as a form of the knowledge graph structure. Specifically, we dynamically construct neighbors and directed edge embeddings based on the head and tail relationships between instances. Then, we devise a knowledge-aware attention mechanism that can update the head node features by learning the joint attention score of each neighbor and edge. Finally, we obtain a graph-level embedding through the global pooling process of the updated head, serving as an implicit representation for the WSI classification. Our end-to-end graph representation learning approach has outperformed the state-of-the-art WSI analysis methods on three TCGA benchmark datasets and in-house test sets. Our code is available at https://github.com/WonderLandxD/WiKG.

Histopathology analysis is the gold standard for medical diagnosis. Accurate classification of whole slide images (WSIs) and region-of-interests (ROIs) localization can assist pathologists in diagnosis. The gigapixel resolution of WSI and the absence of fine-grained annotations make direct classification and analysis challenging. In weakly supervised learning, multiple instance learning (MIL) presents a promising approach for WSI classification. The prevailing strategy is to use attention mechanisms to measure instance importance for classification. However, attention mechanisms fail to capture inter-instance information, and self-attention causes quadratic computational complexity. To address these challenges, we propose AMD-MIL, an agent aggregator with a mask denoise mechanism. The agent token acts as an intermediate variable between the query and key for computing instance importance. Mask and denoising matrices, mapped from agents-aggregated value, dynamically mask low-contribution representations and eliminate noise. AMD-MIL achieves better attention allocation by adjusting feature representations, capturing micro-metastases in cancer, and improving interpretability. Extensive experiments on CAMELYON-16, CAMELYON-17, TCGA-KIDNEY, and TCGA-LUNG show AMD-MIL's superiority over state-of-the-art methods.

In this paper, a dual-polarization dual-beam scanning array antenna based on holographic control theory is proposed. The antenna element integrates four PIN diodes, two in each of the horizonal and vertical directions, to achieve reconfigurable polarization. By varying the states of two PIN diodes in the same direction to change the radiation phase of the element, the simulation and measurement of the 2-element subarray verify that the proposed element is capable of manipulating the radiation pattern. Based on digital coding theory, the phase-adjustable structure integrated with the PIN diodes can act as a holographic element, and the desired object wave can be accurately and stably obtained by modulating the reference wave excited by the feed and the holographic element. The simulation verifies that the antenna array can achieve satisfactory 2D beam scanning. To verify the simulation results, a 2-element subarray prototype and a 6×12 array prototype are fabricated and subjected to measurement. The results of measurement and simulation are in good agreement, which proves the feasibility of the dual-polarization dual-beam scanning antenna system. Due to its low profile, low cost, and easy integration, this antenna system excels in applications in fields such as radar systems and smart antennas.