MICCAI 2025 Papers — Page 8
International Conference on Medical Image Computing and Computer-Assisted Intervention · 1027 papers
PMIL: Prompt enhanced Multimodal Integrative analysis of fMRI combining functional connectivity and temporal Latency
Choi, Hyoungshin (Sungkyunkwan University), Park, Hyunjin (VUNO Inc.)
ClassificationGraph Neural NetworkTransformerPrompt EngineeringVision Language ModelMultimodalityBiomedical DataMagnetic Resonance Imaging
🎯 What it does: A multimodal fusion fMRI analysis framework PMIL is proposed, combining functional connectivity (FC) with cross-time delay structure and intrinsic neural time scale (INT) for the diagnosis of autism spectrum disorder (ASD).
PolarDETR: Enhancing Interpretability in Multi-modal Methods for Jawbone Lesion Detection in CBCT
Yang, Yuxuan (Beijing Jiaotong University), Li, Jupeng (Beijing Jiaotong University)
Object DetectionExplainability and InterpretabilityTransformerImageTextMultimodalityBiomedical DataComputed Tomography
🎯 What it does: This paper presents PolarDETR, a multimodal method that integrates CBCT images with clinical text for the detection of mandibular lesions.
PolyMamba: Spatial-prior Guided Mamba for Polyp Segmentation with High-Frequency Enhancement
Fu, Renyu, Liu, Xinwang (National University Of Defense Technology)
SegmentationTransformerImageBiomedical Data
🎯 What it does: Proposes the PolyMamba framework, which integrates spatial priors generated by Transformer into the Mamba state space model, and designs a dual-gate frequency domain enhancement module to improve polyp segmentation accuracy.
PolypSegTrack: Unified Foundation Model for Colonoscopy Video Analysis
Choudhuri, Anwesa (United Imaging Intelligence), Wu, Ziyan (United Imaging Intelligence)
ClassificationObject DetectionObject TrackingSegmentationTransformerImageVideo
🎯 What it does: We propose PolypSegTrack, a unified foundational model that can achieve polyp detection, segmentation, classification, and unsupervised tracking in colonoscopy videos.
Pose as Clinical Prior: Learning Dual Representations for Scoliosis Screening
Zhou, Zirui (Southern University of Science and Technology), Yu, Shiqi (Southern University of Science and Technology)
ClassificationPose EstimationConvolutional Neural NetworkTransformerVideo
🎯 What it does: A dual representation framework (DRF) has been developed, combining continuous skeleton graphs and discrete pose asymmetric vectors to achieve posture-based scoliosis screening, and a new Scoliosis1K-Pose dataset has been provided.
PR-ENDO: Physically Based Relightable Gaussian Splatting for Endoscopy
Kaleta, Joanna (Sano Centre for Computitional Medicine), Spurek, Przemysław (Sano Centre for Computitional Medicine)
Gaussian SplattingSimultaneous Localization and MappingImageVideo
🎯 What it does: A physics-based relighting model PR-ENDO based on 3D Gaussian Splatting is proposed, which can separate lighting and tissue properties in endoscopic images and supports fast training and real-time rendering.
PRAD: Periapical Radiograph Analysis Dataset and Benchmark Model Development
Zhou, Zhenhuan (Nankai University), Li, Tao (Nankai University)
ClassificationSegmentationConvolutional Neural NetworkImageBenchmark
🎯 What it does: This paper presents a large-scale expert-annotated Periapical Radiograph Dataset (PRAD-10K) and a multi-scale Wavelet Convolution Network (PRNet) specifically designed for this dataset, aimed at pixel-level segmentation and classification of dental PR images.
PRADA: Protecting and Detecting Dataset Abuse for Open-source Medical Dataset
Jang, Jinhyeok (ETRI), Kim, Seong Tae (Technical University of Munich)
Data-Centric LearningConvolutional Neural NetworkBiomedical DataMagnetic Resonance Imaging
🎯 What it does: This paper proposes a framework called PRADA for embedding covert signals in publicly available medical datasets and detecting whether models have been trained on these datasets.
Pre-to-Post Operative MRI Generation with Retrieval-based Visual In-Context Learning
Kang, Bogyeong (Korea University), Kam, Tae-Eui (Korea University)
Image TranslationGenerationRetrievalDiffusion modelImageBiomedical DataMagnetic Resonance Imaging
🎯 What it does: A framework is proposed that utilizes visual context learning and tumor-guided retrieval to generate corresponding postoperative MRI from preoperative brain tumor MRI.
Pre-Trained LLM is a Semantic-Aware and Generalizable Segmentation Booster
Tang, Fenghe (University of Science and Technology of China), Zhou, S. Kevin (Hohai University)
SegmentationConvolutional Neural NetworkLarge Language ModelBiomedical DataMagnetic Resonance Imaging
🎯 What it does: Incorporating frozen pre-trained LLM layers into the CNN segmentation framework to handle visual features and enhance global modeling.
PRECISE-AS: Personalized Reinforcement Learning for Efficient Point-of-Care Echocardiography in Aortic Stenosis Diagnosis
Saadat, Armin (University of British Columbia), Abolmaesumi, Purang (University of British Columbia)
OptimizationComputational EfficiencyTransformerReinforcement LearningVideoBiomedical DataUltrasound
🎯 What it does: A reinforcement learning-based active video collection framework, PRECISE-AS, was designed for personalized selection of cardiac ultrasound videos to diagnose aortic stenosis.
Predicting Alzheimer’s Disease Progression Using a Regression-based Survival Model with Longitudinal Data
Dai, Ling (Xi'an Jiaotong University), Shen, Dinggang (Shanghai United Imaging Intelligence Co., Ltd.)
OptimizationExplainability and InterpretabilityAuto EncoderTabularBiomedical DataMagnetic Resonance ImagingAlzheimer's Disease
🎯 What it does: A regression-based survival analysis framework was designed for predicting the progression of Alzheimer's disease.
Predicting Femoral Head Collapse Risk in Osteonecrosis Using Label Tokenization: A Multi-Modality Survival Analysis Approach
Li, Ganping (Nara Institute of Science and Technology), Sato, Yoshinobu (Nara Institute of Science and Technology)
Convolutional Neural NetworkMultimodalityBiomedical DataMagnetic Resonance Imaging
🎯 What it does: A three-stream deep learning framework is proposed, which combines T1-MRI, ROI annotations, and ONFH grading information to perform survival analysis prediction of the risk of femoral head collapse in the early stages of avascular necrosis using label tokenization and temporal ordering representation.
Predicting Longitudinal Brain Development via Implicit Neural Representations
Dannecker, Maik (Technical University Munich), Rueckert, Daniel (Technical University of Munich)
SegmentationGenerationNeural Radiance FieldImageBiomedical DataMagnetic Resonance Imaging
🎯 What it does: A personalized fetal and neonatal brain development prediction framework based on implicit neural representations (INR) is proposed, which can predict future and past brain structural changes based on a single calibrated scan.
Predicting Radiation Therapy Response based on Dynamic Temporal Feature Difference Fusion from Longitudinal MRI
Hao, Xinyu (Dalian University of Technology), Cong, Fengyu (University of Jyvaskyla)
ClassificationTransformerImageBiomedical DataMagnetic Resonance Imaging
🎯 What it does: A framework is constructed using longitudinal dynamic MRI images to predict pathological complete response (pCR)/treatment response in breast cancer patients after neoadjuvant chemotherapy (or radiotherapy).
PRETI: Patient-Aware Retinal Foundation Model via Metadata-Guided Representation Learning
Lee, Yeonkyung (Yonsei University), Hwang, Seong Jae (Mediwhale)
Representation LearningTransformerAuto EncoderImageBiomedical Data
🎯 What it does: An integrated self-supervised retinal foundation model PRETI was constructed, incorporating patient metadata (age, gender) for efficient learning of representations from color retinal photographs;
Pretraining on Chronic Lung Inflammatory Disease Datasets to Enhance Indeterminant Lung Cancer Classification using Masked Autoencoders
Masquelin, Axel H. P. (Brigham and Women's Hospital), San José Estépar, Raúl (Brigham and Women's Hospital)
ClassificationTransformerAuto EncoderBiomedical DataComputed Tomography
🎯 What it does: Pre-trained a Masked Autoencoder (MAE) on the COPDGene dataset for chronic pneumonia and fine-tuned the pre-trained model for the classification of lung nodule malignancy on the NLST dataset.
PRGNN: Pyramidal Region Graph Neural Network for Region-Based Brain PET Classification
Kim, Daesung (Yonsei University), Yun, Mijin (Dickinson College)
ClassificationConvolutional Neural NetworkGraph Neural NetworkImageBiomedical DataPositron Emission TomographyAlzheimer's Disease
🎯 What it does: This paper proposes a hierarchical regional graph neural network (PRGNN) that extracts multi-scale features through 3D CNN and utilizes brain region ROIs to achieve node embedding for disease classification of PET images.
Prior-guided Prototype Aggregation Learning for Alzheimer’s Disease Diagnosis
Diao, Yueqin (South China University of Technology), Xu, Yanwu (South China University of Technology)
ClassificationConvolutional Neural NetworkLarge Language ModelContrastive LearningImageBiomedical DataMagnetic Resonance ImagingAlzheimer's Disease
🎯 What it does: This paper proposes a prototype aggregation learning framework (PPAL) guided by prior knowledge for the diagnosis of Alzheimer's disease.
Privacy Preserving Chest X-ray Classification in Latent Space with Homomorphically Encrypted Neural Inference
Kim, Jonghun (Sungkyunkwan University), Park, Hyunjin (VUNO Inc.)
ClassificationSafty and PrivacyComputational EfficiencyConvolutional Neural NetworkAuto EncoderGenerative Adversarial NetworkImageBiomedical Data
🎯 What it does: A privacy protection framework for chest X-ray image classification is proposed, which first compresses the image to latent space using VQGAN, and then performs neural network inference on the server side using CKKS homomorphic encryption.
Probabilistic Integration of Renal Cancer Radiology and Pathology Using Graph Neural Networks
Gao, Shangqi (University of Cambridge), Crispin-Ortuzar, Mireia (University of Cambridge)
Graph Neural NetworkBiomedical DataComputed Tomography
🎯 What it does: Proposes a sparse information probabilistic integration of radiomics and pathology to predict overall survival in kidney cancer patients by aggregating spatial features using graph neural networks;
Probabilistic Inverse Consistent Image Registration Using Sparse Bayesian Network
Yang, Shenglong (Shenzhen University), Yang, Xuan (Shenzhen University)
SegmentationGenerationOptimizationAuto EncoderImageBiomedical DataMagnetic Resonance Imaging
🎯 What it does: A probabilistic inverse consistency image registration model based on sparse Bayesian networks and bidirectional variational autoencoders is proposed for cardiac motion estimation, simultaneously outputting the model's Aleatoric and Epistemic uncertainties.
Probabilistic Prior-Guided Anatomical Alignment for MRI Super-Resolution
Luo, Yiwen, Yuan, Yixuan (Southern University of Science and Technology)
RestorationSuper ResolutionTransformerGenerative Adversarial NetworkImageBiomedical DataMagnetic Resonance Imaging
🎯 What it does: This paper studies a probabilistic prior-guided anatomical alignment method called PGASR for super-resolution reconstruction of magnetic resonance imaging.
ProgreSpine: Inherently Explainable Prototypical Regression for Spine Age Estimation
Bazargani, Roozbeh (Prenuvo), Khallaghi, Siavash (Prenuvo)
Explainability and InterpretabilityConvolutional Neural NetworkImageBiomedical DataMagnetic Resonance Imaging
🎯 What it does: A ProgreSpine model based on 3D prototype regression has been developed for age estimation from whole spine T2-weighted MRI;
Progression-Aware Generative Model Enhancing Baseline Visit Prediction of Early Alzheimer’s Disease
Gao, Xingyu (Donghua University), Zhao, Zhe (Donghua University)
GenerationData SynthesisConvolutional Neural NetworkTransformerGenerative Adversarial NetworkImageBiomedical DataPositron Emission TomographyAlzheimer's Disease
🎯 What it does: By combining generative models with Vision Transformer, a small amount of longitudinal 18F-FDG-PET data is used to generate images at future time points, thereby enhancing the early diagnosis of Alzheimer's disease at baseline visits with single time points.
Prolog-Driven Rule-Based Diagnostics with Large Language Models for Precise Clinical Decision Support
Tan, Xiaoyu (Shanghai University of Engineering Science), Qiu, Xihe (Shanghai University of Engineering Science)
TransformerLarge Language ModelPrompt EngineeringTextBiomedical DataChain-of-Thought
🎯 What it does: A clinical decision support system called ProCDS has been developed, which combines Prolog rule reasoning with large language models (LLM) to reduce errors caused by LLM hallucinations and provide transparent and accurate diagnostic suggestions.
Prompt-DAS: Annotation-Efficient Prompt Learning for Domain Adaptive Semantic Segmentation of Electron Microscopy Images
Chen, Jiabao (Huaqiao University), Peng, Jialin (Huaqiao University)
SegmentationDomain AdaptationTransformerPrompt EngineeringContrastive LearningImage
🎯 What it does: We propose Prompt-DAS, a prompt-based Transformer that can perform multi-instance segmentation using sparse point prompts or no prompts in unsupervised, weakly supervised, and interactive domain adaptation tasks for electron microscope image segmentation.
PromptReg: Universal Medical Image Registration via Task Prompt Learning and Domain Knowledge Transfer
Xie, Housheng (Shanghai Jiao Tong University), Zheng, Guoyan (Shanghai Jiao Tong University)
SegmentationDomain AdaptationPrompt EngineeringImageBiomedical DataMagnetic Resonance Imaging
🎯 What it does: This paper proposes PromptReg, which achieves the generalization of a single model in multi-domain medical image registration tasks through prompt learning.
ProTeUS: A Spatio-Temporal Enhanced Ultrasound-Based Framework for Prostate Cancer Detection
Elghareb, Tarek (University of British Columbia), Abolmaesumi, Purang (University of British Columbia)
ClassificationObject DetectionTransformerPrompt EngineeringImageTime SeriesBiomedical DataUltrasound
🎯 What it does: A framework named ProTeUS is proposed and implemented, which integrates B-mode images, RF time-series signals, clinical metadata, and other core information to achieve automatic detection of prostate cancer through foundational models.
Prototype-Based Multiple Instance Learning for Gigapixel Whole Slide Image Classification
Sun, Susu (University of Tübingen), Baumgartner, Christian F. (University of Tübingen)
ClassificationExplainability and InterpretabilityAuto EncoderImageBiomedical Data
🎯 What it does: ProtoMIL has been developed, a multi-instance learning model based on sparse autoencoders that automatically discovers interpretable concepts for whole slide image classification and supports human intervention.
Prototype-Guided and Lightweight Adapters for Inherent Interpretation and Generalisation in Federated Learning
Ofosu Mensah, Samuel (University of Tübingen), Berens, Philipp (University of Tübingen)
Federated LearningExplainability and InterpretabilityConvolutional Neural NetworkImage
🎯 What it does: A framework is proposed that only transmits adapter and prototype parameters in federated learning to address statistical heterogeneity and provide model interpretability.
Prototype-Guided Cross-Modal Knowledge Enhancement for Adaptive Survival Prediction
Liu, Fengchun (Harbin Institute of Technology Shenzhen), Zhang, Yongbing (Harbin Institute of Technology Shenzhen)
ClassificationContrastive LearningMultimodalityBiomedical Data
🎯 What it does: The Prototype-Guided Cross-Modal Knowledge Enhancement (ProSurv) framework is proposed to achieve adaptive survival prediction with unpaired multimodal data.
PSAT: Pediatric Segmentation Approaches via Adult Augmentations and Transfer Learning
Kirscher, Tristan (Institut de cancérologie Strasbourg Europe), Meyer, Philippe (Institut de cancérologie Strasbourg Europe)
SegmentationSupervised Fine-TuningImageBiomedical DataComputed Tomography
🎯 What it does: This paper constructs the PSAT framework and systematically evaluates pediatric CT segmentation strategies based on adult augmentation and transfer learning.
PTCMIL: Multiple Instance Learning via Prompt Token Clustering for Whole Slide Image Analysis
Zhao, Beidi (University Of British Columbia), Li, Xiaoxiao (University Of British Columbia)
ClassificationRepresentation LearningTransformerPrompt EngineeringImage
🎯 What it does: This paper proposes a ViT model called PTCMIL based on visual prompt word clustering for feature aggregation and prediction in multi-instance learning on whole slide images.
pyOpenNFT: an open-source Python framework for ML-based real-time fMRI and EEG-fMRI neurofeedback
Antipushina, Ekaterina (Skolkovo Institute of Science and Technology), Koush, Yury (Samara National Research University)
Biomedical DataMagnetic Resonance Imaging
🎯 What it does: Developed pyOpenNFT, a complete real-time fMRI/EEG-fMRI neurofeedback framework implemented in Python.
Q-space Guided Collaborative Attention Translation Network for Flexible Diffusion-Weighted Images Synthesis
Zhu, Pengli (Hong Kong Polytechnic University), Qiu, Anqi (Hong Kong Polytechnic University)
GenerationData SynthesisGenerative Adversarial NetworkImageBiomedical DataMagnetic Resonance Imaging
🎯 What it does: A Q-space guided collaborative attention translation network (Q-CATN) is proposed to achieve flexible synthesis of multi-shell high angular resolution diffusion imaging (MS-HARDI) based on structural MRI.
Query-Level Alignment for End-to-End Lesion Detection with Human Gaze
Kong, Yan (Nanjing University), Shan, Caifeng (Nanjing University)
Object DetectionExplainability and InterpretabilityTransformerGaussian SplattingImage
🎯 What it does: The study integrates clinical eye movement data into a Transformer-based medical lesion detection model, proposing GAA-DETR to achieve query-level alignment, enhancing detection accuracy and interpretability.
R1Seg-3D: Rethinking Reasoning Segmentation for Medical 3D CTs
Hao, Qin (Xinjiang University), Zhang, Lei (University of Exeter)
SegmentationTransformerLarge Language ModelSupervised Fine-TuningImageBiomedical DataComputed Tomography
🎯 What it does: A three-dimensional medical CT inference segmentation framework R1Seg-3D is proposed, which unifies visual, textual reasoning, and mask decoding in the same latent space.
R2B-WFC Ultrasound Reconstruction: Wavelet Fourier Convolution-based Reconstruction from Radio Frequency to Image
Jeong, Hyunsu (Pohang University of Science and Technology), Kim, Chulhong (Pohang University of Science and Technology)
RestorationGenerationConvolutional Neural NetworkGenerative Adversarial NetworkImageBiomedical DataUltrasound
🎯 What it does: This paper proposes an end-to-end deep learning framework R2B-WFC, which first uses an unsupervised neural Schrödinger bridge (UNSB) to synthesize high-resolution images from low-quality planar wave ultrasound images, and then trains a network with a wavelet-Fourier convolution (WFC) module to directly reconstruct B-mode images from radio frequency (RF) signals.
RadAlign: Advancing Radiology Report Generation with Vision-Language Concept Alignment
Gu, Difei (Rutgers University), Metaxas, Dimitris (Rutgers University)
GenerationRetrievalConvolutional Neural NetworkTransformerLarge Language ModelVision Language ModelContrastive LearningTextMultimodalityBiomedical DataElectronic Health RecordsRetrieval-Augmented Generation
🎯 What it does: The RadAlign framework has been developed to achieve more accurate and interpretable radiology report generation by aligning visual features with medical diagnostic concepts.
Radar-Based Imaging for Sign Language Recognition in Medical Communication
Mineo, Raffaele (University of Catania), Palazzo, Simone (University of Catania)
RecognitionSafty and PrivacyConvolutional Neural NetworkTransformerAuto EncoderMultimodalityBiomedical Data
🎯 What it does: A privacy-preserving medical sign language recognition framework based on 60 GHz millimeter-wave radar is proposed, achieving automatic recognition of 126 medical terms and letters in Italian sign language.
RadGS-Reg: Registering Spine CT with Biplanar X-rays via Joint 3D Radiative Gaussians Reconstruction and 3D/3D Registration
Shen, Ao (Hohai University), Zhou, S. Kevin (Hohai University)
Image TranslationGaussian SplattingImageBiomedical DataComputed Tomography
🎯 What it does: The RadGS-Reg framework is proposed, which reconstructs dual-plane X-ray into Radiative Gaussians (RadGS) and performs 3D/3D registration with preoperative CT, achieving high-precision real-time alignment of vertebral CT/X-ray.
RadioFormer: Integrating Radiologist Inductive Bias for Tumor Classification on Multi-Sequence MR Images
Bai, Xiaoyu (Northwestern Polytechnical University), Xia, Yong (Northwestern Polytechnical University)
ClassificationTransformerImageBiomedical DataMagnetic Resonance Imaging
🎯 What it does: A hierarchical Transformer model named RadioFormer is proposed for tumor classification on multi-sequence MRI data. The model is divided into three layers: single-sequence slice feature extraction → multi-sequence slice information aggregation → inter-slice (volume) information aggregation, simulating the diagnostic process of radiologists and enabling efficient learning on small-scale datasets.
RadiomicsRetrieval: A Customizable Framework for Medical Image Retrieval Using Radiomics Features
Na, Inye (Sungkyunkwan University), Park, Hyunjin (VUNO Inc.)
RetrievalContrastive LearningImageBiomedical DataMagnetic Resonance ImagingComputed Tomography
🎯 What it does: This paper proposes the RadiomicsRetrieval framework, which combines Radiomics features of 3D medical images with deep learning embeddings to achieve tumor-level retrieval based on image, location, or partial features.
RadIR: A Scalable Framework for Multi-Grained Medical Image Retrieval via Radiology Report Mining
Zhang, Tengfei (University of Science and Technology of China), Xie, Weidi (Shanghai AI Laboratory)
RetrievalContrastive LearningImageTextBiomedical DataComputed Tomography
🎯 What it does: A multi-granularity medical image retrieval framework called RadIR based on radiology reports is proposed, and two large-scale retrieval datasets, MIMIC-IR (chest X-ray) and CTRATE-IR (chest CT), are constructed.
RadKAM: Attention-Driven Kolmogorov-Arnold Model for Automatic Radiation-Induced Lymphopenia Prediction by Multimodal Learning
Zhao, Rongchang (Central South University), Li, Shuo (Harbin Institute of Technology)
ClassificationTransformerImageMultimodalityBiomedical DataComputed Tomography
🎯 What it does: This study proposes the RadKAM framework, which utilizes CT images, dose maps, and clinical metadata to predict radiation-induced lymphocyte depletion (RIL) levels without the need for manual ROI segmentation.
RadSAM: Segmenting 3D radiological images with a 2D promptable model
Khlaut, Julien (Raidium), Dancette, Corentin (Raidium)
SegmentationTransformerPrompt EngineeringImageComputed TomographyBenchmark
🎯 What it does: This paper proposes a model named RadSAM, which can perform 3D radiological image segmentation using a single 2D prompt, addressing the issue of traditional methods that require layer-by-layer prompts.
RANDose: A Region-aware Attention Network for Accurate Radiation Dose Prediction
Chowdary, G. Jignesh (Stony Brook University), Yin, Zhaozheng (Stony Brook University)
SegmentationOptimizationConvolutional Neural NetworkBiomedical DataComputed Tomography
🎯 What it does: A regional-aware attention network (RANDose) is proposed for precise prediction of radiation therapy dose distribution.
RAPTOR: Generative AI for Parsing Colorectal Cancer Referrals to Streamline Faster Diagnostic Standard Pathways
Abioye, Sofiat (Birmingham City University), Bilal, Muhammad (Birmingham City University)
TransformerLarge Language ModelSupervised Fine-TuningPrompt EngineeringTextElectronic Health Records
🎯 What it does: This paper develops and evaluates a generative AI-based system called RAPTOR, designed to automatically extract structured information from emergency referral forms for colorectal cancer.
RDMR: Recursive Inference and Representation Disentanglement for Multimodal Large Deformation Registration
Hu, Yibo (Shanghai JiaoTong University), Sun, Jianqi (Shanghai JiaoTong University)
SegmentationDomain AdaptationConvolutional Neural NetworkContrastive LearningMultimodalityBiomedical DataMagnetic Resonance ImagingComputed Tomography
🎯 What it does: A recursive reasoning and representation decoupled multimodal large deformation registration network (RDMR) is proposed to address the challenges of strong intensity differences and large-scale tissue deformations between different modalities.
Real Super-Resolution for Proximal Femur: Enhanced Computation of Structural Bone Metrics from Clinical CTs
Koser, Niklas C. (Kiel University), Glüer, Claus-C. (University Medical Center Hamburg-Eppendorf)
SegmentationSuper ResolutionDiffusion modelGenerative Adversarial NetworkImagePoint CloudBiomedical DataComputed Tomography
🎯 What it does: This study addresses the low-resolution issue of clinical CT images by using a deep learning super-resolution model to reconstruct the proximal femur. An automated point cloud registration method is employed to extract the ROI and calculate bone microstructure indices (BV/TV, Tb.Th, Tb.Sp, Tb.N) to assess the model's potential in predicting bone strength.
Real-Time SLAM-Based Correction and 3D Visualization for Fluorescence Lifetime Imaging
Krishnamoorthy, Murali (Harvard Medical School), Kumar, Anand T. N. (Harvard Medical School)
Depth EstimationOptimizationSimultaneous Localization and MappingImageVideoBiomedical Data
🎯 What it does: This study combines real-time fluorescence lifetime imaging (FLT) with binocular SLAM, utilizing single-gate rapid estimation of FLT and obtaining pixel-level depth information through SLAM, achieving real-time ( >5 Hz) depth-corrected FLT mapping and large-scale 3D surface visualization.
ReCo-I2P: An Incomplete Supervised Lymph Node Segmentation Framework Based on Orthogonal Partial-instance Annotation
Wang, Litingyu (University of Electronic Science and Technology of China), Wang, Guotai (University Of Electronic Science And Technology Of China)
SegmentationConvolutional Neural NetworkImageBiomedical DataComputed Tomography
🎯 What it does: This paper proposes a framework for mediastinal lymph node segmentation based on incomplete supervision, ReCo-I2P, which utilizes extremely sparse orthogonal partial instance annotations (oPIA) for training and achieves high-quality segmentation with only a small amount of annotations.
Recognizing Surgical Phases Anywhere: Few-Shot Test-time Adaptation and Task-graph Guided Refinement
Yuan, Kun (University of Strasbourg), Padoy, Nicolas (University of Strasbourg, CNRS, INSERM, ICube, UMR7357)
RecognitionDomain AdaptationVision Language ModelDiffusion modelContrastive LearningVideoMultimodality
🎯 What it does: Proposes the SPA framework, which utilizes a small amount of labeled data to adapt the medical visual language foundation model spatially, temporally, and during testing, to achieve surgical phase recognition across institutions and surgeries.
Reconsidering Explicit Longitudinal Mammography Alignment for Enhanced Breast Cancer Risk Prediction
Thrun, Solveig (UiT Arctic University of Norway), Kampffmeyer, Michael (UiT Arctic University of Norway)
ClassificationConvolutional Neural NetworkTransformerImageBiomedical Data
🎯 What it does: This paper evaluates explicit alignment methods for longitudinal mammography images, proposing and implementing the first deep learning-based mammography image registration model, MammoRegNet, and comparing the effects of image-level and feature-level alignment in risk prediction tasks.
Reconstructing 3D Hand-Instrument Interaction from a Single 2D Image in Medical Scenes
Xu, Miao (Institute of Automation, Chinese Academy of Sciences), Lei, Zhen (Institute of Automation, Chinese Academy of Sciences)
Pose EstimationConvolutional Neural NetworkImagePoint CloudMesh
🎯 What it does: This paper proposes a method to reconstruct a 3D interaction model of a doctor's hand and medical instruments from a single 2D medical scene image, including steps such as hand initialization, instrument initialization, and a contact point-based interaction module (CPCI).
RedDino: A foundation model for red blood cell analysis
Zedda, Luca (University of Cagliari), Marr, Carsten (Helmholtz Zentrum München - German Research Center for Environmental Health)
ClassificationSegmentationTransformerContrastive LearningImage
🎯 What it does: Developed RedDino - a foundational model for red blood cell analysis based on self-supervised learning, aimed at efficiently and accurately identifying red blood cell morphology;
Reducing Variability of Multiple Instance Learning Methods for Digital Pathology
Mammadov, Ali (LTCI, Télécom Paris, Institut Polytechnique de Paris), Gori, Pietro (Milvue)
ClassificationContrastive LearningImageBiomedical Data
🎯 What it does: In multi-instance learning (MIL) for whole slide image (WSI) classification in digital pathology, the authors propose a strategy based on multi-fidelity model averaging to reduce the high variability of model training results.
RefineNet: Elevating Medical Foundation Models through Quality-Centric Data Curation by MLLM-Annotated Proxy Distillation
Zhang, Ningyi (Macao Polytechnic University), Tan, Tao (Macao Polytechnic University)
Knowledge DistillationData-Centric LearningTransformerLarge Language ModelReinforcement LearningImageTextMultimodalityBiomedical DataMagnetic Resonance Imaging
🎯 What it does: Proposed and implemented RefineNet, a proxy distillation framework based on a multimodal large language model (MLLM) for efficiently screening and enhancing the quality of medical image-text datasets.
RefineSeg: Dual Coarse-to-Fine Learning for Medical Image Segmentation
Du, Anghong (University of Birmingham), Zhang, Le (University of Birmingham)
SegmentationConvolutional Neural NetworkImageBiomedical DataMagnetic Resonance Imaging
🎯 What it does: This paper proposes a weakly supervised medical image segmentation framework called RefineSeg, which utilizes joint learning with positive and negative coarse annotations.
Refining Cervical Cell Classification with Cytological Knowledge and Optimal Attribute Descriptor Matching
Fei, Manman (Shanghai Jiao Tong University), Zhang, Lichi (Shanghai Jiao Tong University)
ClassificationExplainability and InterpretabilityTransformerLarge Language ModelPrompt EngineeringVision Language ModelImageMultimodalityBiomedical DataRetrieval-Augmented Generation
🎯 What it does: A visual-language framework that combines cell morphology attribute description and interpretable attribute matching is proposed for cervical cell classification.
Reflect: Rectified Flows for Efficient Brain Anomaly Correction Transport
Beizaee, Farzad (ÉTS Montreal), Dolz, Jose (CHU Sainte-Justine, University of Montreal)
SegmentationAnomaly DetectionRectified FlowAuto EncoderImageBiomedical DataMagnetic Resonance Imaging
🎯 What it does: Introducing the Reflect framework, using rectified flows for single-step correction of abnormal brain MRI in latent space and achieving unsupervised anomaly detection and localization.
Region-Based Text-Consistent Augmentation for Multimodal Medical Segmentation
Cai, Kunyan, Tan, Tao (Macao Polytechnic University)
SegmentationConvolutional Neural NetworkLarge Language ModelMultimodalityBiomedical DataComputed Tomography
🎯 What it does: A region-based and text-consistent enhancement framework RBTCA is proposed, which maintains semantic consistency between multimodal medical images and text reports during data augmentation.
Register Anything: Estimating “Corresponding Prompts” for Segment Anything Model
Huang, Shiqi (University College London), Hu, Yipeng (King's College London)
Image TranslationSegmentationPrompt EngineeringImageBiomedical DataMagnetic Resonance ImagingComputed Tomography
🎯 What it does: A method called PromptReg is proposed to achieve image registration based on a pre-trained segmentation model by finding corresponding prompt points.
Regression-assisted Classification for CT-based Portal Hypertension Diagnosis
Cai, Wuque (University of Electronic Science and Technology of China), Guo, Daqing (University of Electronic Science and Technology of China)
ClassificationImageBiomedical DataComputed Tomography
🎯 What it does: Proposed the Regression-Assisted Classification (RAC) method and the Boundary-Aware Weighted (BAW) loss based on dynamic weights and boundary adjustments for the diagnosis of portal hypertension (PHT) based on CT images.
RegScore: Scoring Systems for Regression Tasks
Grzeszczyk, Michal K. (Sano Centre for Computational Medicine), Sitek, Arkadiusz (Massachusetts General Hospital)
Explainability and InterpretabilitySupervised Fine-TuningContrastive LearningMultimodalityTabularBiomedical DataMagnetic Resonance Imaging
🎯 What it does: This paper proposes a sparse interpretable regression scoring system called RegScore, and constructs a personalized linear regression (PLR) and personalized scoring (PRS) based on the TIP pre-trained model to integrate tabular and imaging data for predicting pulmonary artery pressure.
Regularized Low-Rank Adaptation for Few-Shot Organ Segmentation
Baklouti, Ghassen (ÉTS Montréal), Ben Ayed, Ismail (ÉTS Montréal)
SegmentationSupervised Fine-TuningImageBiomedical DataComputed Tomography
🎯 What it does: An Adaptive Low-Rank Parameter Tuning (ARENA) method is proposed, which improves the performance of traditional LoRA in medical image few-shot segmentation, capable of automatically adjusting low-rank parameters and enhancing segmentation quality without a validation set.
Reliable and Interpretable Visual Field Progression Prediction with Diffusion Models and Conformal Risk Control
Si, Wenwen (University of Pennsylvania), Al-Aswad, Lama (University of Pennsylvania)
Explainability and InterpretabilityTransformerDiffusion modelBiomedical Data
🎯 What it does: A reliable and interpretable framework for predicting visual progression based on diffusion models and conformal risk control (ABCI) is proposed, along with confidence intervals for typical visual damage patterns.
ReSeg-UNet: A Reconstruction-Guided Optimization Framework for Enhanced Medical Image Segmentation
Li, Lin (Merck & Co., Inc.), Yu, Fei (Hongkong University of Science and Technology)
SegmentationOptimizationConvolutional Neural NetworkImageBiomedical DataMagnetic Resonance ImagingComputed Tomography
🎯 What it does: Enhancing the accuracy of medical image segmentation through a two-stage framework (image reconstruction and three-layer feature alignment),
ResMAP: Restoring MRIs of Mixed Artifacts by Prompt Cascading Retrieval
Tang, Yuxian (ShanghaiTech University), Wang, Qian (United Imaging Intelligence)
RestorationTransformerLarge Language ModelPrompt EngineeringImageBiomedical DataMagnetic Resonance Imaging
🎯 What it does: A cascade framework called ResMAP based on prompt retrieval is proposed for recovering magnetic resonance imaging (MRI) images containing various mixed artifacts;
Resolving the Overlap of Macrovascular and Microvascular Flow Components in Digital Subtraction Angiography for Cerebral Reperfusion Assessment
Wu, Chengchuan (University of Melbourne), Ng, Felix (Melbourne Brain Centre)
RestorationSegmentationImageBiomedical DataMagnetic Resonance Imaging
🎯 What it does: This paper proposes a new neighborhood-based expectation-maximization algorithm that can separate macrovascular and microvascular flow components in digital subtraction angiography images, thereby improving post-stroke reperfusion assessment.
Restyled, Tuning, and Alignment: Taming VLMs for Federated Non-IID Medical Image Analysis
Chen, Shengchao (Shenzhen University), Shu, Ting (Shenzhen University)
ClassificationFederated LearningPrompt EngineeringVision Language ModelImageBiomedical Data
🎯 What it does: In the federated learning environment, the FedTCA framework is proposed to adapt the pre-trained CLIP model for medical image classification, balancing global knowledge and local personalization.
ReSurgSAM2: Referring Segment Anything in Surgical Video via Credible Long-term Tracking
Liu, Haofeng (National University of Singapore), Jin, Yueming (University of Oxford)
Object TrackingSegmentationTransformerVision Language ModelVideoMultimodality
🎯 What it does: A two-stage Surgical Referring Segmentation framework, ReSurgSAM2, is proposed, which first detects the target using text prompts and then performs long-term reliable tracking.
RetFiner: A Vision-Language Refinement Scheme for Retinal Foundation Models
Fecso, Ronald (Medical University of Vienna), Bogunović, Hrvoje (Medical University of Vienna)
ClassificationSegmentationTransformerVision Language ModelContrastive LearningImageTextBiomedical DataElectronic Health Records
🎯 What it does: A RetFiner visual-language refinement scheme is proposed, which refines existing retinal foundation models through self-supervised learning to enhance their semantic understanding and performance on downstream tasks.
Rethinking Multi-view Mammogram Representation Learning via Counterfactual Reasoning with Kolmogorov-Arnold Theorem
Wang, Guoli (Shandong University of Traditional Chinese Medicine), Li, Shuo (Harbin Institute of Technology)
Representation LearningImageBiomedical Data
🎯 What it does: This paper addresses the interference between perspectives in multi-view breast X-ray images, proposing a method to achieve interference removal and effective information utilization through causal graphs and counterfactual reasoning.
RetiDiff: Diffusion-based Synthesis of Retinal OCT Images for Enhanced Segmentation
Li, Sicheng (Zhejiang University), Zhao, Pengpeng (Zhejiang University)
SegmentationGenerationData SynthesisDiffusion modelImageBiomedical Data
🎯 What it does: Utilizing a three-stage diffusion model (DDPM) to synthesize retinal OCT images with segmentation labels, thereby alleviating the problem of scarce labeled data and enhancing segmentation performance.
RetinaLogos: Fine-Grained Synthesis of High-Resolution Retinal Images Through Captions
Ning, Junzhi (Shanghai Artificial Intelligence Laboratory), He, Junjun (Shanghai Artificial Intelligence Laboratory)
GenerationData SynthesisLarge Language ModelVision Language ModelDiffusion modelImageTextBiomedical DataElectronic Health Records
🎯 What it does: This paper presents RetinaLogos, a framework for generating high-resolution retinal images based on text, and constructs the RetinaLogos-1400k dataset, which contains 1.4 million image-caption pairs.
RetSTA: An LLM-Based Approach for Standardizing Clinical Fundus Image Reports
Cai, Jiushen, Li, Huiqi (Beijing Institute of Technology)
TransformerLarge Language ModelSupervised Fine-TuningText
🎯 What it does: A bilingual retinal standard terminology was constructed, and two LLM models (RetSTA-7B-Zero and RetSTA-7B) were trained to achieve automatic standardization of fundus examination reports.
Reverse Imaging for Wide-spectrum Generalization of Cardiac MRI Segmentation
Zhao, Yidong (Delft University of Technology), Tao, Qian (Delft University of Technology)
SegmentationDomain AdaptationDiffusion modelBiomedical DataMagnetic Resonance Imaging
🎯 What it does: This paper proposes the Reverse Imaging method, which estimates the spin characteristics of cardiac MRI images through physics-guided reverse diffusion, enabling interpretable transformations between different sequences and zero-shot generalization.
Revisit the Stability of Vanilla Federated Learning Under Diverse Conditions
Lee, Youngjoon (KAIST), Kang, Joonhyuk (KAIST)
ClassificationFederated LearningTransformerImageBiomedical Data
🎯 What it does: This paper verifies the stability of Vanilla FedAvg under different conditions through federated learning experiments on blood cell and skin lesion classification tasks.
Revisiting 3D Medical Scribble Supervision: Benchmarking Beyond Cardiac Segmentation
Gotkowski, Karol (DKFZ), Isensee, Fabian (DKFZ)
SegmentationConvolutional Neural NetworkBiomedical DataBenchmark
🎯 What it does: This paper evaluates and proposes a highly generalizable bookmark supervision method by constructing a multi-task ScribbleBench benchmark.
Revisiting Automatic Data Curation for Vision Foundation Models in Digital Pathology
Chen, Boqi (University of North Carolina at Chapel Hill), Rätsch, Gunnar (Dept Of Computer Science Eth Zurich)
ClassificationData-Centric LearningTransformerContrastive LearningImage
🎯 What it does: This paper proposes a fully automated tile-level data refinement and sampling framework for visual foundation models (FM) in digital pathology: first, 350M tile embeddings are extracted using a pre-trained FM, and then balanced sampling is achieved using a hierarchical clustering tree; subsequently, a clustering-based hierarchical batch sampling is employed in self-supervised training to mitigate the negative impact of sample imbalance.
Revisiting Masked Image Modeling with Standardized Color Space for Domain Generalized Fundus Photography Classification
Jang, Eojin (DGIST), Park, Sang Hyun (DGIST)
ClassificationDomain AdaptationSupervised Fine-TuningContrastive LearningImage
🎯 What it does: Redesign the Masked Image Modeling (MIM) process by first normalizing color fundus images to a unified color space, then using the normalized images for self-supervised reconstruction, and on this basis, integrating features from both original and transformed images for cross-attention joint learning to enhance domain generalization performance in retinal lesion classification.
RIFNet: Bridging Modalities for Accurate and Detailed Ocular Disease Analysis
Li, Yuqing (Northeastern University), Zaiane, Osmar R. (Amii)
Image TranslationGenerationData SynthesisGenerative Adversarial NetworkImageMultimodality
🎯 What it does: A multi-modal retinal image fusion framework RIFNet has been developed, which first synthesizes pseudo FA images using a dual-stream generative network, and then fuses CFP and FA using Viridis color encoding, ultimately generating high-quality, detail-rich fused images.
Risk Estimation of Knee Osteoarthritis Progression via Predictive Multi-task Modelling from Efficient Diffusion Model using X-ray Images
Butler, David (University of Surrey), Carneiro, Gustavo (University of Surrey)
ClassificationGenerationExplainability and InterpretabilityComputational EfficiencyDiffusion modelAuto EncoderImageBiomedical Data
🎯 What it does: Using an interpretable multi-task model, we predict the future progression of knee osteoarthritis (KL grading) and locate joint anatomical landmarks. The model generates future X-ray images and performs risk estimation by utilizing a diffusion model in the latent space.
RL4Med-DDPO: Reinforcement Learning for Controlled Guidance Towards Diverse Medical Image Generation using Vision-Language Foundation Models
Saremi, Parham (McGill University), Arbel, Tal (McGill University)
GenerationOptimizationReinforcement LearningDiffusion modelImageMultimodalityBiomedical Data
🎯 What it does: To address the semantic alignment issue in text-guided generation of medical images, a framework incorporating reinforcement learning (DDPO) based on Fine-tuned Stable Diffusion is proposed, utilizing a pre-trained attribute classifier as a reward to generate images that are more consistent with the text description.
Robust Deep Learning for Myocardial Scar Segmentation in Cardiac MRI with Noisy Labels
Moafi, Aida (University of Leicester), Mehdipour Ghazi, Mostafa (University of Copenhagen)
Object DetectionSegmentationConvolutional Neural NetworkImageBiomedical DataMagnetic Resonance Imaging
🎯 What it does: This paper proposes a fully automated myocardial scar segmentation pipeline based on YOLO and SAM, utilizing KL loss and extensive data augmentation to overcome issues of semi-automated annotation noise, data heterogeneity, and class imbalance.
Robust Fetal Pose Estimation across Gestational Ages via Cross-Population Augmentation
Diaz, Sebastian (MIT), Adalsteinsson, Elfar (MIT)
Pose EstimationConvolutional Neural NetworkBiomedical DataMagnetic Resonance Imaging
🎯 What it does: A cross-population data augmentation framework has been developed, which incorporates fetus-specific image enhancements (such as fetal position compensation and amniotic fluid reconstruction) during training to achieve generalization of fetal pose estimation across different gestational weeks.
Robust Incomplete-Modality Alignment for Ophthalmic Disease Grading and Diagnosis via Labeled Optimal Transport
Yu, Qinkai (University of Exeter), Meng, Yanda (University Of Exeter)
ClassificationRecognitionOptimizationImageMultimodalityBiomedical Data
🎯 What it does: This paper proposes a multimodal alignment and fusion framework based on Labeled Optimal Transport (Labeled OT) to accurately grade and diagnose ophthalmic diseases even in the absence of retinal photos (fundus) or Optical Coherence Tomography (OCT).
Robust Multimodal Learning for Ophthalmic Disease Grading via Disentangled Representation
Wang, Xinkun (MBZUAI), Razzak, Imran (MBZUAI)
ClassificationKnowledge DistillationRepresentation LearningTransformerImageMultimodalityBiomedical Data
🎯 What it does: A multi-modal learning framework based on essential point and decomposition representation (EDRL) is proposed for grading ophthalmic diseases, addressing the issues of multi-modal redundancy and overlap.
Robust sensitivity control in digital pathology via tile score distribution matching
Pignet, Arthur (Owkin), Olivier, Antoine (Owkin)
ClassificationConvolutional Neural NetworkBiomedical Data
🎯 What it does: A threshold calibration method based on tile score distribution matching (Tile-Score Matching, TSM) is proposed to control the sensitivity of digital pathology WSI binary classification models in multi-instance learning (MIL) models.
Robust Sleep Stage Prediction from Electroencephalogram with Label Noise Using Multimodal Large Language Models
Qiu, Xihe (Shanghai University of Engineering Science), Tan, Xiaoyu (Shanghai University of Engineering Science)
ClassificationTransformerLarge Language ModelPrompt EngineeringMultimodalityTime SeriesBiomedical Data
🎯 What it does: A framework for sleep stage prediction based on a multimodal large language model (MLLM) is proposed, which addresses EEG label noise by using multi-perspective consistency to filter high-quality samples and employs a self-training loop to enhance model robustness.
Robust Topographical Representation for Longitudinal Propagation of Tau Pathology
Yue, Jiaxin (University of Southern California), Shi, Yonggang (University of Southern California)
Anomaly DetectionRepresentation LearningGraph Neural NetworkTime SeriesBiomedical DataPositron Emission TomographyAlzheimer's Disease
🎯 What it does: This paper proposes a multi-resolution Reeb graph (MRG) topological representation for encoding the spatial distribution and propagation dynamics of longitudinal tau PET data, and based on this representation, it achieves joint analysis of subtypes and staging as well as prediction of future tau aggregation.
RRG-DPO: Direct Preference Optimization for Clinically Accurate Radiology Report Generation
Liu, Hong (Eindhoven University of Technology), Wang, Liansheng (Shanghai Changhai Hospital)
GenerationAnomaly DetectionOptimizationLarge Language ModelSupervised Fine-TuningTextBiomedical DataComputed TomographyElectronic Health RecordsRetrieval-Augmented Generation
🎯 What it does: A framework based on Direct Preference Optimization (RRG-DPO) is proposed, which fine-tunes the radiology report generation model using sub-preference loss, significantly reducing hallucination errors and improving clinical accuracy.
RSAD: Region-Specific Anomaly Detection in fMRI for Disease Diagnosis
Sun, Yusong (Shanghai Jiao Tong University), Zhang, Lichi (Shanghai Jiao Tong University)
Anomaly DetectionTransformerAuto EncoderTime SeriesBiomedical DataMagnetic Resonance ImagingAlzheimer's Disease
🎯 What it does: By constructing a self-supervised anomaly detection framework RSAD, brain disease diagnosis is achieved on fMRI data. The core idea is to first use a masked Transformer autoencoder to learn the reconstruction representation of healthy samples, and then utilize region-specific difference scores to determine anomalies in patients.
RT-GAN: Recurrent Temporal GAN for Adding Lightweight Temporal Consistency to Frame-Based Domain Translation Approaches
Mathew, Shawn (Stony Brook University), Kaufman, Arie (Stony Brook University)
Image TranslationGenerationDomain AdaptationRecurrent Neural NetworkGenerative Adversarial NetworkVideoBiomedical Data
🎯 What it does: A lightweight recursive temporal generative adversarial network (RT-GAN) is proposed to add temporal consistency to frame-based unsupervised domain translation models, addressing the inter-frame jitter problem in colonoscopy video analysis.
SABPI-Net: A Novel Structure-Aware Network for Accurate and Domain-Invariant Retinopathy of Prematurity Diagnosis
Chen, Shaobin (Macao Polytechnic University), Sun, Yue (Jinan University)
ClassificationDomain AdaptationTransformerImage
🎯 What it does: A structure-aware bidirectional proxy interaction network (SABPI-Net) was constructed for early ROP diagnosis based on color fundus images, achieving high-accuracy classification on a self-built dataset of over 170K images.
SafeClick: Error-Tolerant Interactive Segmentation of Any Medical Volumes via Hierarchical Expert Consensus
Gao, Yifan (University of Science and Technology of China), Gao, Xin (Chinese Academy of Sciences)
SegmentationTransformerBiomedical DataMagnetic Resonance Imaging
🎯 What it does: Proposes the SafeClick module to enhance the robustness of medical volume segmentation models under imperfect prompts.
SAGCNet: Spatial-Aware Graph Completion Network for Missing Slice Imputation in Population CMR Imaging
Liu, Junkai (University of Birmingham), Zhang, Le (University of Birmingham)
RestorationGraph Neural NetworkTransformerContrastive LearningImageBiomedical DataMagnetic Resonance Imaging
🎯 What it does: A unified framework for missing slice interpolation, SAGCNet, is proposed, which can restore complete cardiac magnetic resonance imaging (CMR) volumes under any missing slice scenario.
SAM2-ProMem: Enhancing Zero-Shot 3D Segmentation with Stochastic Propagation and Memory Search
Wang, Yujie (Shanghai Jiao Tong University), Chen, Boan (Shukun Technology)
SegmentationBiomedical DataMagnetic Resonance ImagingComputed Tomography
🎯 What it does: The study investigates how to transfer SAM2 to 3D medical imaging to achieve zero-shot segmentation of unseen anatomical structures.