Geometry-Aware Superpixel Graph Transformer with Metadata for Skin Lesion Classification

• URL: http://arxiv.org/abs/2606.20390v2
• Date: Thu, 25 Jun 2026 15:23:43 GMT
• Title: Geometry-Aware Superpixel Graph Transformer with Metadata for Skin Lesion Classification
• Abstract summary: Existing CNN/ViT pipelines rely on global or patch-level features and often combine patient metadata via late fusion.<n>We present a novel region-based graph learning framework that explicitly models lesions as graphs of spatially coherent superpixel regions.<n>We show that explicit region-level relational modeling and graph-native multimodal fusion yield consistent gains over the state-of-the-art.
• Score: 8.142478772253932
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Automated skin cancer classification from dermoscopic images remains challenging due to heterogeneous lesion structure, strong intra-class variability, and subtle visual differences between benign and malignant cases. Existing CNN/ViT pipelines typically rely on global or patch-level features and often combine patient metadata via late fusion, which limits spatially grounded multimodal reasoning. We present a novel region-based graph learning framework that explicitly models lesions as graphs of spatially coherent superpixel regions represented as frozen CNN features. To capture fine-grained lesion arrangements, we encode inter-regional geometry as edge attributes and introduce a dedicated metadata context node connected to all regions, providing structured integration of demographic/clinical variables within the same relational space. Node representations are updated using our edge-aware graph transformer followed by attention-driven propagation, and a final graph-level embedding for benign-malignant classification. Experiments on four public benchmarks demonstrate that explicit region-level relational modeling and graph-native multimodal fusion yield consistent gains over the state-of-the-art. Consequently, we establish a new graph-centric perspective in which CNN features are modeled as relational nodes and improved through contextual integration, yielding more expressive and robust classifications.

Related papers

• An Attention-Guided Global and Local Fusion Framework for Lesion-Focused Image Classification [0.0]
  This study hypothesizes that adaptive fusion of global contextual information and lesion-focused local information can improve classification performance.<n>We propose a three-branch, attention-guided deep learning framework to improve feature attribution, interpretability, and classification reliability.
  arXiv  Detail & Related papers  (2026-09-04T06:36:06Z)
• Gated Multi-Graph Fusion via Graph Attention Networks for Alzheimer's Disease Detection [52.36198268369384]
  Spontaneous speech is a vital non-invasive biomarker for Alzheimer's Disease (AD)<n>We propose a Multi-View Gated Graph Attention Network that transcribes audio via Automatic Speech Recognition (ASR) to construct semantic, dependency, and co-occurrence graphs.
  arXiv  Detail & Related papers  (2026-06-30T06:15:59Z)
• TFFM: Topology-Aware Feature Fusion Module via Latent Graph Reasoning for Retinal Vessel Segmentation [3.903438631397551]
  We introduce a topology-aware framework engineered to maintain vascular connectivity.<n>Our architecture fuses a Topological Feature Fusion Module (TFFM) that maps local feature representations into a latent graph space.<n>We drive the learning process with a hybrid objective function, coupling Tversky loss for class imbalance with soft clDice loss to explicitly penalize topological disconnects.
  arXiv  Detail & Related papers  (2026-01-27T03:09:10Z)
• Integrating Multi-scale and Multi-filtration Topological Features for Medical Image Classification [20.820287362872975]
  Deep neural networks have shown remarkable performance in medical image classification.<n>We propose a new topology-guided classification framework that extracts multi-scale and multi-filtration persistent topological features.<n>Our approach enhances the model's capacity to recognize complex anatomical structures.
  arXiv  Detail & Related papers  (2025-12-08T06:02:02Z)
• A Graph-Based Framework for Interpretable Whole Slide Image Analysis [86.37618055724441]
  We develop a framework that transforms whole-slide images into biologically-informed graph representations.<n>Our approach builds graph nodes from tissue regions that respect natural structures, not arbitrary grids.<n>We demonstrate strong performance on challenging cancer staging and survival prediction tasks.
  arXiv  Detail & Related papers  (2025-03-14T20:15:04Z)
• Histopathology Whole Slide Image Analysis with Heterogeneous Graph Representation Learning [78.49090351193269]
  We propose a novel graph-based framework to leverage the inter-relationships among different types of nuclei for WSI analysis. Specifically, we formulate the WSI as a heterogeneous graph with "nucleus-type" attribute to each node and a semantic attribute similarity to each edge. Our framework outperforms the state-of-the-art methods with considerable margins on various tasks.
  arXiv  Detail & Related papers  (2023-07-09T14:43:40Z)
• How GNNs Facilitate CNNs in Mining Geometric Information from Large-Scale Medical Images [2.2699159408903484]
  We propose a fusion framework for enhancing the global image-level representation captured by convolutional neural networks (CNNs) We evaluate our fusion strategies on histology datasets curated from large patient cohorts of colorectal and gastric cancers.
  arXiv  Detail & Related papers  (2022-06-15T15:27:48Z)
• Mixed Graph Contrastive Network for Semi-Supervised Node Classification [63.924129159538076]
  We propose a novel graph contrastive learning method, termed Mixed Graph Contrastive Network (MGCN)<n>In our method, we improve the discriminative capability of the latent embeddings by an unperturbed augmentation strategy and a correlation reduction mechanism.<n>By combining the two settings, we extract rich supervision information from both the abundant nodes and the rare yet valuable labeled nodes for discriminative representation learning.
  arXiv  Detail & Related papers  (2022-06-06T14:26:34Z)
• BI-GCN: Boundary-Aware Input-Dependent Graph Convolution Network for Biomedical Image Segmentation [21.912509900254364]
  We apply graph convolution into the segmentation task and propose an improved textitLaplacian. Our method outperforms the state-of-the-art approaches on the segmentation of polyps in colonoscopy images and of the optic disc and optic cup in colour fundus images.
  arXiv  Detail & Related papers  (2021-10-27T21:12:27Z)
• Spatial-spectral Hyperspectral Image Classification via Multiple Random Anchor Graphs Ensemble Learning [88.60285937702304]
  This paper proposes a novel spatial-spectral HSI classification method via multiple random anchor graphs ensemble learning (RAGE) Firstly, the local binary pattern is adopted to extract the more descriptive features on each selected band, which preserves local structures and subtle changes of a region. Secondly, the adaptive neighbors assignment is introduced in the construction of anchor graph, to reduce the computational complexity.
  arXiv  Detail & Related papers  (2021-03-25T09:31:41Z)
• Hyperbolic Graph Embedding with Enhanced Semi-Implicit Variational Inference [48.63194907060615]
  We build off of semi-implicit graph variational auto-encoders to capture higher-order statistics in a low-dimensional graph latent representation. We incorporate hyperbolic geometry in the latent space through a Poincare embedding to efficiently represent graphs exhibiting hierarchical structure.
  arXiv  Detail & Related papers  (2020-10-31T05:48:34Z)
• Multi-Level Graph Convolutional Network with Automatic Graph Learning for Hyperspectral Image Classification [63.56018768401328]
  We propose a Multi-level Graph Convolutional Network (GCN) with Automatic Graph Learning method (MGCN-AGL) for HSI classification. By employing attention mechanism to characterize the importance among spatially neighboring regions, the most relevant information can be adaptively incorporated to make decisions. Our MGCN-AGL encodes the long range dependencies among image regions based on the expressive representations that have been produced at local level.
  arXiv  Detail & Related papers  (2020-09-19T09:26:20Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.