Progressive Experience Fusion for Multi-Task World Model Control in Endovascular Navigation

• URL: http://arxiv.org/abs/2608.18647v1
• Date: Wed, 19 Aug 2026 07:51:50 GMT
• Title: Progressive Experience Fusion for Multi-Task World Model Control in Endovascular Navigation
• Abstract summary: This study investigates Progressive Experience Fusion (PEF) to train a multi-stage TD-MPC2 controller.<n>We additionally evaluate a plan that changes the Model Predictive Path Integral planning horizon using residual action-sequence dispersion, and fine-tuning in a patient-specific simulation.
• Score: 37.517747868704184
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Autonomous endovascular navigation could support the delivery of mechanical thrombectomy to underserved areas, but controllers must navigate long, multi-stage paths across varying vascular anatomies. This study investigates Progressive Experience Fusion (PEF) to train a multi-task TD-MPC2 controller. We additionally evaluate a heuristic that changes the Model Predictive Path Integral planning horizon using residual action-sequence dispersion, and fine-tuning in a patient-specific simulation. Across five subtasks in ten known training anatomies with held-out targets, PEF achieved a mean success rate of 74%, compared with 37% for Soft Actor-Critic (p < 0.001) and 65% for base TD-MPC2 (p = 0.053). A PEF controller with adaptive-horizon planning trained on 30 vasculatures achieved a mean success rate of 90% in ten held-out vasculatures. The PEF agent successfully transferred to an unseen in vitro stroke patient vasculature under fluoroscopy, achieving a mean path ratio improvement from 63% to 80% with fine-tuning (p < 0.001), following 40x103 fine-tuning steps (corresponding to approximately 107 min of clinical inter-hospital transfer time). This work represents a proof of concept for multi-vasculature training and patient-specific adaptation, while further validation is required before clinical deployment.

Related papers

• A Clinically Validated Foundation Model for Comprehensive Lung Pathology Interpretation [36.95302007351382]
  PulmoFoundation is a prospectively validated, randomized controlled trial (RCT)-evaluated foundation model for comprehensive lung pathology assessment.<n>Our model achieves clinical-grade performance in core diagnostic tasks across biopsy, frozen section, and surgical resection slides.
  arXiv  Detail & Related papers  (2026-05-25T14:04:56Z)
• Anatomical Landmark-Guided Deep Reinforcement Learning for Autonomous Gastric Navigation [9.172107354380117]
  Wireless capsule endoscopy (WCE) enables painless visualization of the gastrointestinal tract.<n>We propose a transferable, anatomical landmarkguided deep reinforcement learning framework for autonomous gastric navigation.
  arXiv  Detail & Related papers  (2026-05-08T03:11:21Z)
• A Breast Vision Pathology Foundation Model for Real-world Clinical Utility [65.57568187389113]
  We present bfBRAVE, a breast-adaptive pathology foundation model developed and evaluated using a total resource of 101,638 breast wholeslide images.<n>BRAVE supported practical roles in the clinical workflow, including safe exclusion of low-risk cases from routine AI-assisted second-review rescue of initially missed positives and prioritization of cases for further assessment.
  arXiv  Detail & Related papers  (2026-05-06T07:44:39Z)
• Toward Safe Autonomous Robotic Endovascular Interventions using World Models [34.54099310430464]
  We investigate a world-model-based framework for autonomous endovascular navigation built on TD-MPC2.<n>We evaluate a TD-MPC2 agent trained on multiple navigation across hold out patient-specific vasculatures.<n>We benchmark its performance against the state-of-the-art Soft Actor-Critic (SAC) algorithm agent.
  arXiv  Detail & Related papers  (2026-04-22T03:32:18Z)
• Transparent Early ICU Mortality Prediction with Clinical Transformer and Per-Case Modality Attribution [42.85462513661566]
  We present a lightweight, transparent multimodal ensemble that fuses physiological time-series measurements with unstructured clinical notes from the first 48 hours of an ICU stay.<n>A logistic regression model combines predictions from two modality-specific models: a bidirectional LSTM for vitals and a finetuned ClinicalModernBERT transformer for notes.<n>On the MIMIC-III benchmark, our late-fusion ensemble improves discrimination over the best single model while maintaining well-calibrated predictions.
  arXiv  Detail & Related papers  (2025-11-19T20:11:49Z)
• Cancer-Net PCa-MultiSeg: Multimodal Enhancement of Prostate Cancer Lesion Segmentation Using Synthetic Correlated Diffusion Imaging [55.62977326180104]
  Current deep learning approaches for prostate cancer lesion segmentation achieve limited performance.<n>We investigate synthetic correlated diffusion imaging (CDI$s$) as an enhancement to standard diffusion-based protocols.<n>Our results establish validated integration pathways for CDI$s$ as a practical drop-in enhancement for PCa lesion segmentation tasks.
  arXiv  Detail & Related papers  (2025-11-11T04:16:12Z)
• Fluence Map Prediction with Deep Learning: A Transformer-based Approach [4.362977319915193]
  This study presents a deep learning framework that accelerates fluence map generation while maintaining clinical quality.<n>An end-to-end 3D Swin-UNETR network was trained to predict nine-beam fluence maps directly from volumetric CT images and anatomical contours.<n>The proposed model achieved an average R2 of 0.95 +/- 0.02, MAE of 0.035 +/- 0.008, and gamma passing rate of 85 +/- 10 percent (3 percent / 3 mm) on the test set.
  arXiv  Detail & Related papers  (2025-11-10T19:24:14Z)
• World Model for AI Autonomous Navigation in Mechanical Thrombectomy [39.51808126417126]
  We propose a world model for autonomous endovascular navigation using TD-MPC2, a model-based RL algorithm.<n>We trained a single RL agent across multiple endovascular navigation tasks in ten real patient vasculatures, comparing performance against the state-of-the-art Soft Actor-Critic (SAC) method.<n>Results indicate that TD-MPC2 significantly outperforms SAC in multi-task learning, achieving a 65% mean success rate compared to SAC's 37%, with notable improvements in path ratio.
  arXiv  Detail & Related papers  (2025-09-29T21:21:30Z)

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.