MICCAI 2025 Papers — Page 11
International Conference on Medical Image Computing and Computer-Assisted Intervention · 1027 papers
VesselGPT: Autoregressive Modeling of Vascular Geometry
Feldman, Paula (Consejo Nacional de Investigaciones Científicas y Técnicas), Iarussi, Emmanuel (Consejo Nacional de Investigaciones Científicas y Técnicas)
GenerationData SynthesisTransformerAuto EncoderMeshGraph
🎯 What it does: A framework for autoregressive generation of vascular geometry, called VesselGPT, is proposed. It first learns a discrete vocabulary of node attributes using VQ-VAE, then generates a sequence of nodes using GPT-2, and finally decodes the sequence into a three-dimensional vascular network.
VesselSDF: Distance Field Priors for Vascular Network Reconstruction
Esposito, Salvatore (University of Edinburgh), Mac Aodha, Oisin (University of Edinburgh)
SegmentationConvolutional Neural NetworkImageBiomedical DataComputed Tomography
🎯 What it does: Proposes the VesselSDF two-stage framework, which reconstructs the vascular network in sparse CT slices using signed distance fields (SDF).
VesselVerse: A Dataset and Collaborative Framework for Vessel Annotation
Falcetta, Daniele (EURECOM), Zuluaga, Maria A. (EURECOM)
Object DetectionSegmentationSupervised Fine-TuningImageBiomedical DataMagnetic Resonance ImagingComputed Tomography
🎯 What it does: This paper presents the VesselVerse dataset and collaborative framework, summarizing 950 brain vascular images from IXI, TubeTK, and TopCoW, and provides multi-expert (both human and model) annotations with traceable version control.
Vision-Amplified Semantic Entropy for Hallucination Detection in Medical Visual Question Answering
Liao, Zehui (Northwestern Polytechnical University), Xia, Yong (Northwestern Polytechnical University)
Anomaly DetectionTransformerContrastive LearningImageTextMultimodalityMagnetic Resonance Imaging
🎯 What it does: Proposes the Vision-Amplified Semantic Entropy (VASE) method for detecting model hallucination answers in medical visual question answering (VQA).
VisNet: A Human Visual System Inspired Lightweight Dual-Path Network for Medical Images Denoising
Yue, Hailin (Central South University), Wang, Jianxin (Institute of Guizhou Aerospace Measuring and Testing Technology)
RestorationConvolutional Neural NetworkImageBiomedical DataMagnetic Resonance ImagingComputed Tomography
🎯 What it does: This paper proposes a lightweight dual-path network called VisNet, which draws inspiration from the human visual system's radiating cell pathways, primary visual cortex, and dorsal/ventral pathways. It designs dual-path multi-scale perception, edge detection and shape adaptation, and spatial semantic extraction modules to achieve efficient denoising of CT, X-ray, and MRI images.
ViTAL-CT: Vision Transformers for High-Risk Plaque Classification in Coronary CTA
Le, Anjie (University of Cambridge), Huang, Yuan (University of Cambridge)
ClassificationTransformerSupervised Fine-TuningImageBiomedical DataComputed Tomography
🎯 What it does: A segmentation-free Vision Transformer-based ViTAL-CT framework is proposed to classify high-risk plaques directly on the CTA plaque cross-sections aligned with the coronary artery centerline.
ViTexNet: Vision-Text Guided Dynamic Convolution Network for Medical Image Segmentation
Bhardwaj, Rahul (Indian Institute of Technology Guwahati), Neog, Debanga Raj (Indian Institute of Technology Guwahati)
SegmentationConvolutional Neural NetworkVision Language ModelImageTextMultimodalityBiomedical DataComputed Tomography
🎯 What it does: A multimodal medical image segmentation network ViTexNet based on visual-text dynamic convolution is proposed, utilizing text information to guide convolution for efficient segmentation.
VMRA-MaR: An Asymmetry-Aware Temporal Framework for Longitudinal Breast Cancer Risk Prediction
Sun, Zijun (University of Bologna), Kampffmeyer, Michael (UiT Arctic University of Norway)
Recurrent Neural NetworkTime SeriesBiomedical Data
🎯 What it does: This study investigates a breast cancer risk prediction framework that integrates temporal information and symmetry analysis.
VoxelOpt: Voxel-Adaptive Message Passing for Discrete Optimization in Deformable Abdominal CT Registration
Zhang, Hang (Cornell University), Liu, Min (Hunan University)
OptimizationImageComputed Tomography
🎯 What it does: This paper presents VoxelOpt, a discrete optimization framework based on voxel adaptive information transfer for elastic registration of abdominal CT images.
VQ-SCD: Vector Quantization Meets Unknown Scan Condition Self-supervised Low-Dose CT Denoising
Su, Bo (Wuhan University), Lu, Zhouxian (Wuhan University)
RestorationTransformerDiffusion modelImageComputed Tomography
🎯 What it does: A self-supervised low-dose CT denoising framework VQ-SCD is proposed, which utilizes positive-dose CT training to achieve low-dose image denoising under unknown scanning conditions.
VT-SNN: Variable Time-step Spiking Neural Network Based on Uncertainty Measure and Its Application in Brain Disease Diagnosis
Rao, Haonan (Nantong University), Huang, Jiashuang (Nantong University)
ClassificationAnomaly DetectionComputational EfficiencySpiking Neural NetworkTransformerImageBiomedical DataMagnetic Resonance Imaging
🎯 What it does: A variable time-step pulse neural network (VT-SNN) is proposed, which can dynamically terminate inference based on sample uncertainty, improving the efficiency and accuracy of brain magnetic resonance imaging diagnosis.
WASABI: A Metric for Evaluating Morphometric Plausibility of Synthetic Brain MRIs
Jafrasteh, Bahram (Weill Cornell Medicine), Zhao, Qingyu (Weill Cornell Medicine)
GenerationData SynthesisDiffusion modelImageBiomedical DataMagnetic Resonance ImagingAlzheimer's Disease
🎯 What it does: A new metric named WASABI is proposed and validated to assess the morphological reliability of generated brain MR images, and it is used to compare the performance of various advanced generative models.
WaveFormer: A 3D Transformer with Wavelet-Driven Feature Representation for Efficient Medical Image Segmentation
Al Hasan, Md Mahfuz (University of Florida), Forghani, Reza (University of Florida)
SegmentationTransformerImageBiomedical DataMagnetic Resonance ImagingComputed Tomography
🎯 What it does: WaveFormer is proposed, a 3D Transformer that utilizes Discrete Wavelet Transform (DWT) to perform self-attention at low frequencies and uses Inverse Discrete Wavelet Transform (IDWT) to recover details at high frequencies, aimed at medical image segmentation.
Wavelet-driven Decoupling and Physics-informed Mapping Network for Accelerated Multi-parametric MR Imaging
Dan, Ruilong (ShanghaiTech University), Shen, Dinggang (Shanghai United Imaging Intelligence Co., Ltd.)
Convolutional Neural NetworkBiomedical DataMagnetic Resonance Imaging
🎯 What it does: An end-to-end accelerated multi-parameter MRI reconstruction and parameter mapping network based on wavelet separation and physical constraints (WDPM-Net) is proposed.
WDNet: A Novel Wavelet-guided Hierarchical Diffusion Network for Multi-Target Segmentation in Colonoscopy Images
He, Dongdong (Harbin Institute of Technology), Fu, Yili (Harbin Institute of Technology)
SegmentationConvolutional Neural NetworkImageStochastic Differential Equation
🎯 What it does: A hierarchical network WDNet based on wavelet decomposition and diffusion reconstruction is proposed for precise segmentation of polyps and surgical instruments in colonoscopy images.
Weakly Semi-Supervised Cervical Lesion Cell Detection via Twin-Memory Augmented Multiple Instance Learning
Fei, Manman (Shanghai Jiao Tong University), Zhang, Lichi (Shanghai Jiao Tong University)
Object DetectionSupervised Fine-TuningImageBiomedical Data
🎯 What it does: Proposes the Twin-MIL framework, which utilizes weak slide-level labels to enhance cell-level pathological detection;
Weakly-Supervised 2D/3D Image Registration via Differentiable X-ray Rendering and ROI Segmentation
Cui, Yuxin, Min, Zhe (Shandong University)
SegmentationPose EstimationOptimizationConvolutional Neural NetworkSupervised Fine-TuningImageBiomedical DataComputed Tomography
🎯 What it does: A weakly supervised 2D/3D image registration framework is proposed, utilizing ROI segmentation information in X-ray projections and rendered images, and employing a differentiable Dice loss to achieve joint learning and optimization of registration and segmentation.
Weighted Stratification in Multi-Label Contrastive Learning for Long-Tailed Medical Image Classification
Lin, Ying-Chih (National Yang Ming Chiao Tung University), Chen, Yong-Sheng (National Yang Ming Chiao Tung University)
ClassificationContrastive LearningImageBiomedical Data
🎯 What it does: This paper proposes a multi-label contrastive learning framework named ws-MulSupCon to address the long-tail distribution and comorbidity issues in medical image classification.
WiD-PET: PET Image Reconstruction from Low-Dose Data Using a Wavelet-Informed Diffusion Model with Fast Inference
Lyu, Qingcheng (University of Sydney), Zhou, Luping (University of Sydney)
RestorationConvolutional Neural NetworkDiffusion modelImagePositron Emission Tomography
🎯 What it does: A fast diffusion model based on wavelet transform, WiD-PET, is proposed for the reconstruction of low-dose PET images, aimed at recovering standard dose images from ultra-low dose data.
World Model for AI Autonomous Navigation in Mechanical Thrombectomy
Robertshaw, Harry (King's College London), Booth, Thomas C. (King's College London)
Autonomous DrivingOptimizationRobotic IntelligenceRecurrent Neural NetworkReinforcement LearningWorld ModelMeshBiomedical DataComputed Tomography
🎯 What it does: A reinforcement learning method based on world models (TD-MPC2) is proposed and implemented for multi-task vascular navigation during mechanical thrombectomy (MT) procedures.
WSI-Agents: A Collaborative Multi-Agent System for Multi-Modal Whole Slide Image Analysis
Lyu, Xinheng (Shenzhen University), Shen, Linlin (Shenzhen University)
ClassificationSegmentationGenerationRetrievalTransformerLarge Language ModelAgentic AIVision Language ModelImageMultimodalityBiomedical DataBenchmark
🎯 What it does: A collaborative multi-agent system (WSI-Agents) has been constructed for the analysis of multi-modal whole slide images (WSI), integrating task allocation, internal consistency verification, external knowledge validation, and summarization modules to achieve multi-task diagnosis, report generation, and other functions.
X-SiT: Inherently Interpretable Surface Vision Transformers for Dementia Diagnosis
Bongratz, Fabian (Technical University of Munich), Wachinger, Christian (Technical University of Munich)
ClassificationExplainability and InterpretabilityTransformerMeshBiomedical DataAlzheimer's Disease
🎯 What it does: An interpretable model X-SiT based on Surface Vision Transformer is designed for the diagnosis of Alzheimer's disease and frontotemporal dementia, achieving interpretability of the decision-making process through a prototype surface patch decoder.
XFMamba: Cross-Fusion Mamba for Multi-View Medical Image Classification
Zheng, Xiaoyu (Queen Mary University of London), Slabaugh, Greg (Queen Mary University of London)
ClassificationConvolutional Neural NetworkImageBiomedical DataMagnetic Resonance Imaging
🎯 What it does: A network called XFMamba based on Mamba is proposed for multi-view medical image classification, and a two-stage cross-view fusion mechanism is designed.
XOCT: Enhancing OCT to OCTA Translation via Cross-Dimensional Supervised Multi-Scale Feature Learning
Khosravi, Pooya (University of California, Irvine), Xie, Xiaohui (University of California, Irvine)
Image TranslationGenerationGenerative Adversarial NetworkImage
🎯 What it does: Proposed and implemented the XOCT framework, achieving deep learning translation from OCT images to OCTA images while preserving the continuity and details of blood vessels during the translation process.
You Can Detect It: Fetal Biometric Estimation Using Ellipse Detection
Zhang, Hongyuan (HKISI-CAS), Wu, Songxiong (Shenzhen University)
Object DetectionSegmentationOptimizationConvolutional Neural NetworkImageBiomedical DataUltrasound
🎯 What it does: A single-stage ellipse detection network, EllipseDet, is proposed to directly estimate the cardiothoracic ratio (CTR) in the fetal four-chamber view through ellipse regression, avoiding the traditional segmentation post-processing workflow.
Your other Left! Vision-Language Models Fail to Identify Relative Positions in Medical Images
Wolf, Daniel (Ulm University), Ropinski, Timo (Ulm University)
TransformerLarge Language ModelVision Language ModelImageBiomedical DataComputed TomographyBenchmark
🎯 What it does: This study investigates the ability of existing visual-language models to recognize relative positions in medical images and proposes the MIRP benchmark dataset.
µ2 Tokenizer: Differentiable Multi-Scale Multi-Modal Tokenizer for Radiology Report Generation
Li, Siyou (Queen Mary University of London), Zhang, Le (University of Birmingham)
GenerationTransformerLarge Language ModelSupervised Fine-TuningImageTextMultimodalityComputed Tomography
🎯 What it does: A differentiable multi-scale multi-modal tokenizer µ 2 Tokenizer is proposed, combined with a large language model µ 2 LLM to achieve CT image report generation.