Graph Neural Networks for Surgical Scene Segmentation

A recent study analyzed the multi-modal approach in the Vision Transformer (EVA-ViT) image encoder combined with LlaMA and ChatGPT large language models (LLMs) to address hallucination issues and enhance disease detection in chest X-ray images. The research utilized the NIH Chest X-ray dataset, comparing image-based and text-based retrieval-augmented generation (RAG) methods, revealing that text-based RAG effectively mitigates hallucinations while image-based RAG improves prediction confidence.

A novel deep learning framework named TIMM-ProRS has been introduced to enhance the prognosis and risk stratification of diabetic retinopathy (DR), a condition that threatens the vision of millions worldwide. This framework integrates Vision Transformer, Convolutional Neural Network, and Graph Neural Network technologies, utilizing both retinal images and temporal biomarkers to achieve a high accuracy rate of 97.8% across multiple datasets.

A new framework named InfGraND has been introduced to facilitate Influence-guided Knowledge Distillation from Graph Neural Networks (GNNs) to Multi-Layer Perceptrons (MLPs). This framework aims to enhance the efficiency of MLPs by prioritizing structurally influential nodes in the graph, addressing challenges faced by traditional GNNs in low-latency and resource-constrained environments.

A new framework named GADPN has been proposed to enhance Graph Neural Networks (GNNs) by refining graph topology through low-rank denoising and generalized structural perturbation, addressing issues of noise and missing links in graph-structured data.

Recent research highlights the potential of Subgraph Graph Neural Networks (GNNs) for node classification, addressing the limitations of traditional node-centric approaches that suffer from high computational costs and scalability issues. The proposed SubGND framework aims to enhance efficiency while maintaining classification accuracy through innovative techniques like differentiated zero-padding and Ego-Alter subgraph representation.

A novel framework named Directed Homophily-aware Graph Neural Network (DHGNN) has been introduced to address the challenges faced by traditional Graph Neural Networks (GNNs) in generalizing to heterophilic neighborhoods and in processing directed graphs. DHGNN incorporates homophily-aware and direction-sensitive components, utilizing a resettable gating mechanism and a noise-tolerant fusion module to enhance performance.