GALAR-TemporalNet v2: Anatomy-Guided Dual-Branch Temporal Classification with Bidirectional Mamba and Dual-Graph GCN for Video Capsule Endoscopy -- after competition results

• URL: http://arxiv.org/abs/2605.22209v1
• Date: Thu, 21 May 2026 09:15:43 GMT
• Title: GALAR-TemporalNet v2: Anatomy-Guided Dual-Branch Temporal Classification with Bidirectional Mamba and Dual-Graph GCN for Video Capsule Endoscopy -- after competition results
• Abstract summary: We present a hierarchical temporal model that addresses three core challenges: extreme class imbalance, long-range temporal dependencies, and pathology--anatomy entanglement.<n>Our architecture combines windowed self-attention for local modeling, a Dual-Graph GCN for global frame relationships, and Bidirectional Mamba for selective boundary context encoding.
• Score: 2.9421642252347624
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Video Capsule Endoscopy (VCE) poses a challenging multi-label temporal classification problem, requiring simultaneous localization of 8 anatomical regions and detection of 9 pathological findings across tens of thousands of frames. We present GALAR-TemporalNet v2, a hierarchical temporal model that addresses three core challenges: extreme class imbalance, long-range temporal dependencies, and pathology--anatomy entanglement. Our architecture combines windowed self-attention for local modeling, a Dual-Graph GCN for global frame relationships, and Bidirectional Mamba for selective boundary context encoding. A novel anatomy prototype residual pathway decouples pathological deviation signals from normal organ appearance, and a frame-level GCN skip connection stabilizes training of visually confusable rare classes. The competition version, GALAR-TemporalNet, achieved an overall mAP@0.5 of 0.2644 and mAP@0.95 of 0.2353 on the RARE-VISION test set. Following the competition, the redesigned GALAR-TemporalNet v2 -- incorporating a restructured pathology branch, refined loss functions, and extended post-processing -- improved these results to mAP@0.5 of 0.3409 and mAP@0.95 of 0.3333.

Related papers

• Disentangling Co-Occurring Retinal Pathologies with Saliency-Guided Sparse Expert Routing [3.332219204498619]
  We propose a novel architecture that resolves this via sparse conditional computation.<n>We show that expert routing aligns with localized lesions and geometrically maps co-occurring cases.
  arXiv  Detail & Related papers  (2026-08-10T15:43:43Z)
• MedMamba: Multi-View State Space Models with Adaptive Graph Learning for Medical Time Series Classification [60.41879345743941]
  We propose MedMamba, an end-to-end architecture that integrates state space models with domain-specific inductive biases.<n>MedMamba achieves state-of-the-art performance while maintaining linear computational complexity.
  arXiv  Detail & Related papers  (2026-05-24T09:23:39Z)
• Differentiable latent structure discovery for interpretable forecasting in clinical time series [38.473526928012724]
  We introduce StructGP, a continuous-time multi-task Gaussian process that couples differentiable structure learning to uncover a sparse, ordered directed acyclic graph (DAG) of inter-variable dependencies while preserving principled uncertainty.<n>We propose LP-StructGP, which augments StructGP with latent pathways-shared, temporally shifted trajectories inferred via subject-specific coupling filters and a softmax gating mechanism-to capture cross-patient progression patterns.
  arXiv  Detail & Related papers  (2026-04-30T14:59:50Z)
• Training-inference input alignment outweighs framework choice in longitudinal retinal image prediction [3.0602131532925436]
  Recent methods have moved toward increasing generative complexity, but the conditions under which this complexity is necessary remain unclear.<n>We propose that generative complexity should match the entropy of a predictable task's conditional posterior component, with training-inference input alignment required in all regimes.<n>We validated this framework on a fundus autofluorescence dataset by contrasting five configurations.
  arXiv  Detail & Related papers  (2026-04-18T10:28:00Z)
• Neuro-Oracle: A Trajectory-Aware Agentic RAG Framework for Interpretable Epilepsy Surgical Prognosis [4.726777092009554]
  Trajectory-based classifiers achieve AUC values between 0.834 and 0.905, compared with 0.793 for a single-timepoint ResNet-50 baseline.<n>We propose emphNeuro-Oracle, a three-stage framework that: (i) distils pre-to-post-operative MRI changes into a compact 512-dimensional trajectory vector using a 3D Siamese contrastive encoder; (ii) retrieves historically similar surgical trajectories from a population archive via nearest-neighbour search; and (iii) synthesises a natural-language prognosis grounded in the retrieved evidence using a quantized
  arXiv  Detail & Related papers  (2026-04-10T21:47:25Z)
• VISTA: Validation-Guided Integration of Spatial and Temporal Foundation Models with Anatomical Decoding for Rare-Pathology VCE Event Detection [39.50945484737978]
  We formulate the RARE-VISION task as a metric-aligned event detection problem instead of a purely frame-wise classification task.<n>Our framework combines two complementary backbones, EndoFM-LV for local temporal context and DINOv3 ViT-L/16 for strong frame-level visual semantics.<n> validation ablations indicate that complementary backbones, validation-guided fusion, and anatomy-aware temporal decoding all contribute to event-level performance.
  arXiv  Detail & Related papers  (2026-03-18T23:01:12Z)
• How Much Temporal Modeling is Enough? A Systematic Study of Hybrid CNN-RNN Architectures for Multi-Label ECG Classification [1.8119312186036625]
  We evaluate the necessity and clinical justification of deep and stacked recurrent architectures for ECG classification.<n>A CNN integrated with a single BiLSTM layer achieves the most favorable trade-off between predictive performance and model complexity.<n>These findings suggest that architectural alignment with the intrinsic temporal structure of ECG signals, rather than increased recurrent depth, is a key determinant of robust performance.
  arXiv  Detail & Related papers  (2026-01-25T17:29:13Z)
• AttentiveGRUAE: An Attention-Based GRU Autoencoder for Temporal Clustering and Behavioral Characterization of Depression from Wearable Data [46.262619407930266]
  We present AttentiveGRUAE, a novel attention-based gated recurrent unit (GRU) autoencoder designed for temporal clustering and prediction of outcome from longitudinal wearable data.<n>We evaluate AttentiveGRUAE on longitudinal sleep data from 372 participants (GLOBEM 2018-2019)<n>It demonstrates superior performance over baseline clustering, domain-aligned self-supervised, and ablated models in both clustering quality and depression classification.
  arXiv  Detail & Related papers  (2025-10-02T20:52:16Z)
• Topology-Constrained Learning for Efficient Laparoscopic Liver Landmark Detection [46.2391319253146]
  Liver landmarks provide crucial anatomical guidance to the surgeon during laparoscopic liver surgery.<n>TopoNet is a novel topology-constrained learning framework for laparoscopic liver landmark detection.<n>Our framework adopts a snake-CNN dual-path encoder to simultaneously capture detailed RGB texture information and depth-informed topological structures.
  arXiv  Detail & Related papers  (2025-07-01T07:35:36Z)
• A topology-preserving three-stage framework for fully-connected coronary artery extraction [39.47027832777158]
  Coronary artery extraction is a crucial prerequisite for computer-aided diagnosis of coronary artery disease.<n>We propose a topology-preserving three-stage framework for fully-connected coronary artery extraction.<n>This framework includes vessel segmentation, centerline reconnection, and missing vessel reconstruction.
  arXiv  Detail & Related papers  (2025-04-02T11:04:44Z)
• KaLDeX: Kalman Filter based Linear Deformable Cross Attention for Retina Vessel Segmentation [46.57880203321858]
  We propose a novel network (KaLDeX) for vascular segmentation leveraging a Kalman filter based linear deformable cross attention (LDCA) module. Our approach is based on two key components: Kalman filter (KF) based linear deformable convolution (LD) and cross-attention (CA) modules. The proposed method is evaluated on retinal fundus image datasets (DRIVE, CHASE_BD1, and STARE) as well as the 3mm and 6mm of the OCTA-500 dataset.
  arXiv  Detail & Related papers  (2024-10-28T16:00:42Z)
• Deep Learning for Vascular Segmentation and Applications in Phase Contrast Tomography Imaging [33.23991248643144]
  We present a thorough literature review, highlighting the state of machine learning techniques across diverse organs. Our goal is to provide a foundation on the topic and identify a robust baseline model for application to vascular segmentation in a new imaging modality. HiP CT enables 3D imaging of complete organs at an unprecedented resolution of ca. 20mm per voxel.
  arXiv  Detail & Related papers  (2023-11-22T11:15:38Z)
• Weakly Supervised 3D Classification of Chest CT using Aggregated Multi-Resolution Deep Segmentation Features [5.938730586521215]
  Weakly supervised disease classification of CT imaging suffers from poor localization owing to case-level annotations. We propose a medical classifier that leverages semantic structural concepts learned via multi-resolution segmentation feature maps.
  arXiv  Detail & Related papers  (2020-10-31T00:16:53Z)

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.