MICCAI 2025 Papers — Page 7
International Conference on Medical Image Computing and Computer-Assisted Intervention · 1027 papers
Multi-Modal Graph-Based Machine Learning for Predicting Surgical Outcome in Epilepsy Patients
Aharonyan, Artur Arturi (Sheikh Zayed Institute for Pediatric Surgical Innovation, Children's National Hospital), Anwar, Syed Muhammad (Children's National Hospital)
ClassificationGraph Neural NetworkMultimodalityBiomedical DataMagnetic Resonance Imaging
🎯 What it does: This paper proposes a multimodal spatiotemporal graph neural network that utilizes sEEG (with thalamic electrodes) and structural MRI data to predict the binary outcome of reduced seizure frequency after epilepsy surgery.
Multi-modal Knowledge Decomposition based Online Distillation for Biomarker Prediction in Breast Cancer Histopathology
Zhang, Qibin, Xu, Hongming (Dalian University Of Technology)
ClassificationKnowledge DistillationImageMultimodalityBiomedical Data
🎯 What it does: This paper proposes an online distillation method based on multimodal knowledge decomposition to predict IHC biomarkers from H&E stained whole slide images of breast cancer;
Multi-modal MRI Translation via Evidential Regression and Distribution Calibration
Liu, Jiyao, Zhuang, Xiahai (Fudan University)
Image TranslationGenerationDomain AdaptationGenerative Adversarial NetworkMultimodalityBiomedical DataMagnetic Resonance Imaging
🎯 What it does: This paper proposes a multimodal MRI translation framework based on evidence regression and distribution calibration, which can estimate and utilize uncertainty during the synthesis process, achieving reliable image generation and enhancing robustness across centers.
Multi-Modal Progressive Fusion for ASD Screening Using Smartphone Video
Zhong, Wenqi, Zhang, Dingwen (Northwestern Polytechnical University)
ClassificationRecognitionConvolutional Neural NetworkTransformerVideoMultimodality
🎯 What it does: Developed a multimodal ASD screening framework based on smartphone videos, extracting eye movement, head posture, and emotional features for deep integration.
Multi-modal Representations for Fine-grained Multi-label Critical View of Safety Recognition
Baby, Britty (University of Strasbourg, CNRS, INSERM, ICube, UMR7357), Padoy, Nicolas (University of Strasbourg, CNRS, INSERM, ICube, UMR7357)
RecognitionConvolutional Neural NetworkLarge Language ModelPrompt EngineeringContrastive LearningImageMultimodalityBiomedical Data
🎯 What it does: This paper proposes CVSAdaptNet, which utilizes multimodal text prompts for the multi-label automatic recognition of critical safety views (CVS) in laparoscopic cholecystectomy, avoiding reliance on spatial annotations.
Multi-scale Attention-based Multiple Instance Learning for Breast Cancer Diagnosis
Mourão, Mariana (Instituto Superior Técnico), Silveira, Margarida (Instituto Superior Técnico)
ClassificationObject DetectionConvolutional Neural NetworkTransformerImage
🎯 What it does: This study investigates a multi-scale attention-based multi-instance learning framework for weakly supervised classification and localization of breast cancer, capable of extracting multi-scale features and dynamically fusing information from different scales under single-scale input.
Multi-Sensory Cognitive Computing for Learning Population-level Brain Connectivity
Soussia, Mayssa (University of Sousse), Rekik, Islem (University of Sousse)
ClassificationRecognitionComputational EfficiencyRecurrent Neural NetworkSupervised Fine-TuningMultimodalityBiomedical DataMagnetic Resonance Imaging
🎯 What it does: A multi-sensory cognitive computing framework mCOCO based on stochastic Reservoir Computing is proposed, which learns population-level connectivity templates from BOLD signals and introduces multimodal inputs to enhance cognitive abilities.
Multi-Spatial Granger Causality Features Fusion Network for Alzheimer’s Disease Classification
Song, Zhiwei (Beijing Normal University), Guo, Xiaojuan (Beijing Normal University)
ClassificationConvolutional Neural NetworkGraph Neural NetworkContrastive LearningImageBiomedical DataMagnetic Resonance ImagingAlzheimer's Disease
🎯 What it does: Using a multi-spatial Granger causality feature fusion network for Alzheimer's disease classification based on sMRI
Multi-subject Orthogonal Sparse Matrix Decomposition Method for Extracting Individual Brain Functional Networks
He, Xingyu (Shanxi University), Du, Yuhui (Shanxi University)
Biomedical DataMagnetic Resonance Imaging
🎯 What it does: A multi-agent quasi-orthogonal sparse matrix decomposition method is proposed to simultaneously extract group-level functional networks and individual-level functional networks without group-level analysis or complex initialization, while maintaining cross-subject correspondence.
Multi-task Screening for Cervical Diseases via Feature Routing and Asymmetric Distillation
Jiang, Haotian (ShanghaiTech University), Wang, Qian (United Imaging Intelligence)
ClassificationKnowledge DistillationTransformerImageBiomedical Data
🎯 What it does: A unified multi-task early cervical disease screening framework MECDS is proposed, which can simultaneously handle three tasks: cervical cancer, candidiasis infection, and atypical cells, and can expand to new tasks without retraining the entire model.
Multi-Tracer Uptake Correction for PET-MR via Aligned-Feature Guidance and Multi-scale Pixel-adaptive Routing
Zhong, Aocheng (ShanghaiTech University), Wang, Qian (United Imaging Intelligence)
TransformerMixture of ExpertsAuto EncoderContrastive LearningBiomedical DataMagnetic Resonance ImagingComputed TomographyPositron Emission Tomography
🎯 What it does: This paper proposes and implements a multi-tracker PET-MR quantitative correction framework, achieving quantitative consistency between PET-MR and PET-CT through CT feature representation, MR-to-CT feature alignment, and multi-scale pixel routing under a three-stage training process.
Multi-Tube-Voltage vBMD Measurement via Dual-Branch Frequency Balancing and Asymmetric Channel Attention
Zhang, Mengze (United Imaging Intelligence Co Ltd), Qian, Zhen (United Imaging Intelligence Co Ltd)
Convolutional Neural NetworkImageBiomedical DataComputed Tomography
🎯 What it does: A lightweight multi-frequency deep network is proposed for achieving phantom-less voxel bone mineral density (vBMD) measurement under various CT tube voltages.
Multi-view Graph Contrastive Learning with Dynamic Self-aware and Cross-sample Topology Augmentation for Brain Disorder Diagnosis
Zhang, Hao (Nanjing Forestry University), Zhang, Li (Nanjing Forestry University)
ClassificationAnomaly DetectionGraph Neural NetworkContrastive LearningGraphBiomedical DataMagnetic Resonance Imaging
🎯 What it does: Construct a rs-fMRI brain network and propose a multi-view graph contrastive learning framework MGCL-DA, utilizing dynamic adaptation and cross-sample topology enhancement for automatic diagnosis of brain diseases.
Multifrequency Neural Network-based Wave Inversion in MR Elastography
Bustin, Héloïse (Deutsches Herzzentrum der Charité), Hennemuth, Anja (Deutsches Herzzentrum der Charité)
TransformerBiomedical DataMagnetic Resonance Imaging
🎯 What it does: A multi-frequency neural network-based MRE waveform inversion method called MF-ElastoNet is proposed, which can directly utilize the wavefields collected at multiple frequencies for elastic modulus reconstruction.
Multimodal Fusion Network with Distribution-based Tumor-Marker Imputation for Multi-Origin Metastatic Cervical Lymphadenopathy Classification
Li, Rui (Sun Yat-sen University), Lu, Yao (Sun Yat-sen University)
ClassificationRecognitionConvolutional Neural NetworkGenerative Adversarial NetworkMultimodalityBiomedical DataUltrasound
🎯 What it does: Aiming at the classification of primary tumors in multi-source metastatic cervical lymphoma (CLA), a multimodal fusion network (MDFN) is proposed, which integrates B-mode ultrasound (BUS), color Doppler flow imaging (CDFI), and distribution-based tumor marker (TM) imputation.
Multimodal Hypergraph Guide Learning for Non-Invasive ccRCC Survival Prediction
Yan, Jielong (Tsinghua University), Gao, Yue (Tsinghua University)
ClassificationGraph Neural NetworkMultimodalityBiomedical DataComputed Tomography
🎯 What it does: A multimodal hypergraph-guided learning framework is proposed for non-invasive prediction of survival time in patients with ccRCC (clear cell renal cell carcinoma).
Multimodal Imputation of Imaging-derived Phenotypes from Genomic and Blood-based Biomarkers Enhances Common Disease Discovery
Zhang, Haoyang (Fudan University), Wang, Chengyan (Fudan University)
Data-Centric LearningDrug DiscoveryMultimodalityBiomedical DataMagnetic Resonance Imaging
🎯 What it does: A multimodal model based on genomic SNPs and blood biomarkers was constructed to predict 260 brain and heart MRI-derived phenotypes (IDP), and it was applied to a large-scale UK biobank lacking imaging data to achieve 'imputation' of IDP and conduct disease association and prediction studies.
Multimodal Prompt Sequence Learning for Interactive Segmentation of Vascular Structures
Lim, Jongsoo (Kookmin University), Lee, Soochahn (Kookmin University)
SegmentationTransformerPrompt EngineeringMultimodalityBiomedical Data
🎯 What it does: This paper proposes an interactive vascular structure segmentation method based on multi-modal (text + point) prompt sequence learning, which can progressively improve segmentation results through user interaction in various medical images.
Multiscale Graph and Multi-Step Cross-Frame Mamba for Myocarditis Lesion Segmentation
Yu, Chengjin, Pu, Cailing (Shaoxing People's Hospital)
SegmentationGraph Neural NetworkImageBiomedical DataMagnetic Resonance Imaging
🎯 What it does: This paper proposes a framework called MG-Mamba for the segmentation of myocarditis lesions in non-contrast Cine-MRI images.
Multistage Alignment and Fusion for Multimodal Multiclass Alzheimer’s Disease Diagnosis
Huang, Shuo (University of Southern California), Shi, Yonggang (University of Southern California)
ClassificationTransformerContrastive LearningMultimodalityTabularBiomedical DataMagnetic Resonance ImagingPositron Emission TomographyFibre Orientation DistributionAlzheimer's Disease
🎯 What it does: This paper proposes a multi-stage alignment and fusion framework that utilizes multimodal imaging (T1-MRI, Tau-PET, FOD) and clinical tables (age, gender, MoCA score) to achieve a three-class diagnosis of Alzheimer's disease (CN, MCI, AD).
MultiTransAD: Cross-Sequence Translation-Driven Anomaly Detection in Multi-Sequence Brain MRI
Zhang, Qi (Shanghai Jiao Tong University), Sun, Jianqi (Shanghai JiaoTong University)
Anomaly DetectionTransformerImageBiomedical DataMagnetic Resonance Imaging
🎯 What it does: A multi-sequence MRI unsupervised anomaly detection framework called MultiTransAD is proposed, which directly quantifies anomalies using cross-sequence translation errors.
Multiview Feature Fusion and Contrastive Learning for Drug-Target Interaction Prediction
Zeng, Xiaoting (Shenzhen University), Lei, Baiying (Shenzhen University)
Drug DiscoveryGraph Neural NetworkContrastive LearningGraphBiomedical Data
🎯 What it does: A drug-target interaction prediction framework MFCL-DTI based on multi-view feature fusion and contrastive learning is proposed, integrating features from sequence, neighbor, and meta-path perspectives.
MurreNet: Modeling Holistic Multimodal Interactions Between Histopathology and Genomic Profiles for Survival Prediction
Liu, Mingxin (Nanjing University of Information Science and Technology), Xu, Jun (Heilongjiang University)
ClassificationTransformerMultimodalityBiomedical Data
🎯 What it does: The MurreNet framework is proposed, achieving global interaction between pathological images and genomic information for cancer survival prediction through multimodal representation decoupling and deep orthogonal fusion.
Mutual Information Regularization for Fairness-aware Deep Imaging Representations
Sadri, Amir Reza (Case Western Reserve University), Viswanath, Satish E. (Case Western Reserve University)
ClassificationRepresentation LearningTransformerImageBiomedical DataMagnetic Resonance Imaging
🎯 What it does: This paper proposes a fair deep image representation learning method based on mutual information regularization (FaMI), aimed at eliminating the model's dependence on sensitive attributes such as race or gender and enhancing fairness;
MvHo-IB: Multi-View Higher-Order Information Bottleneck for Brain Disorder Diagnosis
Zhang, Kunyu (Zhengzhou University), Yu, Shujian (Zhengzhou University)
ClassificationExplainability and InterpretabilityConvolutional Neural NetworkGraph Neural NetworkBiomedical DataMagnetic Resonance ImagingAlzheimer's Disease
🎯 What it does: A multi-view information bottleneck framework MvHo-IB is constructed, which jointly learns the pairwise connections and triple interaction information of brain regions, and compresses irrelevant redundancy through the information bottleneck.
MVP-LLMs: Optimizing Intervention Timing and Subsequent Decision Support for Mechanical Ventilation Parameter Control Using Large Language Models
Hao, Teqi (Shanghai University of Engineering Science), Qiu, Xihe (Shanghai University of Engineering Science)
OptimizationReinforcement Learning from Human FeedbackTransformerLarge Language ModelReinforcement LearningPrompt EngineeringTime SeriesSequentialBiomedical DataElectronic Health RecordsChain-of-Thought
🎯 What it does: Utilizing large language models (LLM) to simulate the mechanical ventilation parameter adjustment process of ICU doctors, the task is divided into two phases: optimal stopping and subsequent decision-making, enhancing decision quality through scheduled prompts and chain-of-thought reasoning.
Navigational Bronchoscopy in Critical Care via End-to-End Pose Regression
Mackute, Emile (University of Edinburgh), Khadem, Mohsen (University of Edinburgh)
Pose EstimationTransformerVideo
🎯 What it does: This paper proposes a navigation bronchoscope system suitable for critically ill patients on mechanical ventilation, utilizing a vision transformer to achieve end-to-end pose regression and closed-loop correction through visual landmarks, enabling real-time guidance without preoperative imaging or external sensors.
NAVIUS: Navigated Augmented Reality Visualization for Ureteroscopic Surgery
Acar, Ayberk (Vanderbilt University), Wu, Jie Ying (Florida International University)
ImageBiomedical DataComputed Tomography
🎯 What it does: The NAVIUS system is proposed and implemented, which displays a 3D model of the renal pelvis system generated from preoperative CT in real-time on HoloLens 2, along with the EM tracking position of the ureteroscope, helping surgeons to explore the renal pelvis more comprehensively.
Neighborhood-Consistent Binary Transformation for Domain-Invariant Chest X-ray Diagnosis
Liu, Zelong (Wuhan University), Xu, Yongchao (Wuhan University)
ClassificationDomain AdaptationAuto EncoderImageBiomedical Data
🎯 What it does: An unsupervised domain-invariant chest X-ray diagnosis framework based on Neighborhood Consistent Binarization Transformation (NCBT) and Intermediate Domain Style Preserving Autoencoder (IDSP-AE) is proposed;
NeRF-based CBCT Reconstruction needs Normalization and Initialization
Xu, Zhuowei (University of Science and Technology of China), Zhou, S. Kevin (Hohai University)
RestorationOptimizationNeural Radiance FieldGaussian SplattingBiomedical DataComputed Tomography
🎯 What it does: This paper addresses the issue of local-global training mismatch between the hash encoder and MLP in NeRF-based CBCT reconstruction, proposing two solutions: normalized hash encoder and mapping consistency initialization, which significantly enhance training stability and convergence speed.
NERO: Explainable Out-of-Distribution Detection with Neuron-level Relevance in Gastrointestinal Imaging
Chhetri, Anju (NepAl Applied Mathematics and Informatics Institute for research), Bhattarai, Binod (University of Aberdeen)
Anomaly DetectionExplainability and InterpretabilityConvolutional Neural NetworkTransformerImageBiomedical Data
🎯 What it does: A post-hoc OOD detection method called NERO is proposed, which determines whether a sample is OOD by clustering neuron-level correlation centers and calculating correlation distances, while also providing interpretability.
Neural Proteomics Fields for Super-resolved Spatial Proteomics Prediction
Zhao, Bokai (University of Chinese Academy of Sciences), Jiang, Tianzi (University of Chinese Academy of Sciences)
GenerationSuper ResolutionConvolutional Neural NetworkTransformerNeural Radiance FieldBiomedical DataBenchmark
🎯 What it does: A neural network framework called Neural Proteomics Fields (NPF) for super-resolution prediction of spatial proteomics (seq-SP) is proposed, and a Pseudo-Visium SP benchmark dataset is constructed.
Neuro-AMS: Neuro-informed Age-aware and Medical Knowledge-integrated Strategy for Diagnosis of Multiple Brain Disorders
Zhang, Zhenguo (ShanghaiTech University), Shen, Dinggang (Shanghai United Imaging Intelligence Co., Ltd.)
ClassificationContrastive LearningImageBiomedical DataMagnetic Resonance ImagingAlzheimer's Disease
🎯 What it does: This paper proposes the Neuro-AMS framework, which achieves automatic diagnosis of various brain diseases through semantic assistance and age-aware strategies.
NeuroXVocal: Detection and Explanation of Alzheimer’s Disease through Non-invasive Analysis of Picture-prompted Speech
Ntampakis, Nikolaos (International Hellenic University), Argyriou, Vasileios (Kingston University London)
ClassificationExplainability and InterpretabilityTransformerTextMultimodalityAlzheimer's DiseaseRetrieval-Augmented GenerationAudio
🎯 What it does: Developed NeuroXVocal, an end-to-end interpretable Alzheimer's disease detection system that integrates multimodal audio, text, and speech embeddings.
New Multimodal Similarity Measure for Image Registration via Modeling Local Functional Dependence with Linear Combination of Learned Basis Functions
Honkamaa, Joel (Aalto University), Marttinen, Pekka (Aalto University)
OptimizationConvolutional Neural NetworkImageMultimodalityBiomedical DataMagnetic Resonance ImagingComputed Tomography
🎯 What it does: A similarity measure for multimodal image registration based on local functional dependence is proposed, and an efficient implementation with a closed-form solution is achieved through learning a linear basis function model.
New Multiple Sclerosis Lesion Segmentation via Calibrated Inter-patch Blending
Ye, Jin (Monash University), Cai, Jianfei (Monash University)
SegmentationConvolutional Neural NetworkImageBiomedical DataMagnetic Resonance Imaging
🎯 What it does: A multi-scale fusion framework based on calibrated patch weights (CIB) is proposed for 3D segmentation of new multiple sclerosis lesions.
NIMOSEF: Neural implicit motion and segmentation functions
Banus, Jaume, Richiardi, Jonas (University of Lausanne)
RestorationSegmentationConvolutional Neural NetworkAuto EncoderBiomedical DataMagnetic Resonance Imaging
🎯 What it does: Developed the NIMOSEF framework, which jointly achieves high-resolution segmentation, image reconstruction, and displacement field estimation for cardiac MRI.
No More Sliding Window: Efficient 3D Medical Image Segmentation with Differentiable Top-K Patch Sampling
Jeon, Young Seok (National University of Singapore), Feng, Mengling (National University of Singapore)
SegmentationComputational EfficiencyConvolutional Neural NetworkBiomedical DataComputed Tomography
🎯 What it does: Proposes the NMSW framework, which utilizes differentiable Top-K sampling to select only a small number of high-importance patches during inference, combining low-resolution global predictions and patch-level predictions to complete 3D medical image segmentation, eliminating the high time and memory consumption issues of traditional sliding window inference.
Noise-Controllable Complex-Valued Diffusion Model for k-Space Data of Hyperpolarized 129Xe Lung MRI Generation
Han, Linxuan (Huazhong University of Science and Technology), Zhou, Xin (Huazhong University of Science and Technology)
GenerationData SynthesisConvolutional Neural NetworkDiffusion modelImageBiomedical DataMagnetic Resonance Imaging
🎯 What it does: This paper proposes a noise-controllable complex-valued diffusion model (NC-CDM) for directly generating k-space data of hyperpolarized 129Xe lung MRI to alleviate the issue of sample scarcity.
Noise-Robust Tuning of SAM for Domain Generalized Ultrasound Image Segmentation
Wei, Zhikai (Wuhan University), Xu, Yongchao (Wuhan University)
SegmentationDomain AdaptationTransformerSupervised Fine-TuningImageUltrasound
🎯 What it does: This paper presents Nora, a noise-robust fine-tuning framework for ultrasound image segmentation, which achieves cross-domain generalization by injecting adaptive noise into the feature space of SAM and constructing instance-aware prompts.
Noisy Label Refinement Based on Discrete Diffusion Process in 3D Ossicle Segmentation
Fan, Linqian (Tsinghua University), Yin, Hongxia (Tsinghua University)
SegmentationDiffusion modelImageBiomedical DataComputed Tomography
🎯 What it does: A volume label refinement framework VDDR based on discrete diffusion processes is proposed, combined with a Dilating & Selecting mechanism, to refine noise annotations for three types of middle ear ossicles.
Non-Invasive TB Detection using Acoustic and Semantic Features from Cough Sounds
Akhter, Yasmeena (Indian Institute of Technology), Singh, Richa (Indian Institute of Technology)
ClassificationLarge Language ModelAudio
🎯 What it does: A dual-stream deep learning architecture called AcouSem-AFNet is proposed for non-invasive tuberculosis detection through acoustic and semantic features.
Non-Salient Object Segmentation in Medical Images via Pre-trained Multi-Granularity Masked Autoencoders
Zhang, Bin (Anhui University), Wang, Rui (Beijing AnZhen Hospital)
SegmentationAuto EncoderImageBiomedical DataComputed Tomography
🎯 What it does: A multi-granularity mask autoencoder (MG-MAE) is proposed for the segmentation of non-significant objects in medical images, achieving multi-scale feature learning through global pixel reconstruction and local HOG feature refinement.
NQNN: Noise-aware Quantum Neural Networks for Medical Image Classification
Rahman, Maqsudur (Boise State University), Zhuang, Jun (Boise State University)
ClassificationImageBiomedical DataMagnetic Resonance Imaging
🎯 What it does: A noise-aware quantum neural network (NQNN) is proposed to improve robustness in multi-class medical image classification tasks with label noise.
Oblique Genomics Mixture of Experts: Prediction of Brain Disorder With Aging-Related Changes of Brain’s Structural Connectivity Under Genomic Influences
Lyu, Yanjun (University of Texas at Arlington), Zhu, Dajiang (University of Georgia)
ClassificationTransformerMixture of ExpertsContrastive LearningTime SeriesBiomedical DataDiffusion Tensor ImagingAlzheimer's Disease
🎯 What it does: A sparse expert Transformer (OG-MoE) that combines time series structural connectivity (SC) with genomic information was constructed and trained, achieving predictions of changes in brain region structural connectivity associated with Alzheimer's disease (AD) or mild cognitive impairment (MCI) through self-supervised pre-training and multi-task learning.
Occlusion-free 4D Gaussians for Open Surgery Videos Using Multi-Camera Shadowless Lamps
Kato, Yuna (Keio University), Isogawa, Mariko (Keio University)
RestorationSegmentationGenerationData SynthesisGaussian SplattingVideo
🎯 What it does: Using multi-camera shadowless lamps (McSL) to record open surgery videos, 4D Gaussian scattering technology is employed to generate editable, unobstructed 3D videos that support arbitrary viewpoint switching.
ODES: Online Domain Adaptation with Expert Guidance for Medical Image Segmentation
Islam, Md Shazid (University of California, Riverside), Roy-Chowdhury, Amit K. (University of California, Riverside)
SegmentationDomain AdaptationImageBiomedical DataMagnetic Resonance ImagingOrdinary Differential Equation
🎯 What it does: An online medical image segmentation domain adaptation framework ODES is proposed, which combines expert-guided active learning to achieve single forward and backward updates.
OFF-CLIP: Improving Normal Detection Confidence in Radiology CLIP with Simple Off-Diagonal Term Auto-Adjustment
Park, Junhyun (DGIST), Hwang, Minho (DGIST)
ClassificationAnomaly DetectionTransformerPrompt EngineeringContrastive LearningImageTextBiomedical DataElectronic Health Records
🎯 What it does: This paper proposes OFF-CLIP, an improved method to enhance the accuracy of normal case detection in radiology CLIP models by incorporating off-diagonal loss and text filtering.
Omni-Fusion of Spatial and Spectral for Hyperspectral Image Segmentation
Zhang, Qing (East China Normal University), Wang, Yan (East China Normal University)
SegmentationTransformerImage
🎯 What it does: A global spatial-spectral fusion network called Omni-Fuse is proposed for microscopic hyperspectral image segmentation.
On the Interplay of Human-AI Alignment, Fairness, and Performance Trade-offs in Medical Imaging
Luo, Haozhe (University of Bern), Reyes, Mauricio (University of Bern)
ClassificationTransformerVision Language ModelImageBiomedical Data
🎯 What it does: The system evaluates the fairness and performance of chest X-ray disease classification models under Human-AI Alignment.
One For All: A Unified Approach to Classification and Self-Explanation
Naouar, Mehdi (University of Freiburg), Kalweit, Maria (University of Freiburg)
ClassificationExplainability and InterpretabilityTransformerImageBiomedical DataMagnetic Resonance Imaging
🎯 What it does: This paper proposes a One For All (OFA) method that unifies image classification and self-explanation (based on Shapley distribution approximation) for training within the same model, avoiding the use of post-hoc explainers.
One-shot active learning for vessel segmentation
Falcetta, Daniele (EURECOM), Zuluaga, Maria A. (EURECOM)
SegmentationContrastive LearningImageBiomedical DataMagnetic Resonance Imaging
🎯 What it does: A one-round active learning framework V-DiSNet is designed, which uses dictionary learning and k-means to generate a sparse latent space, selecting a small number of vascular image patches for expert annotation to train a brain vascular segmentation model.
OpenPath: Open-Set Active Learning for Pathology Image Classification via Pre-trained Vision-Language Models
Zhong, Lanfeng (University of Electronic Science and Technology of China), Wang, Guotai (University Of Electronic Science And Technology Of China)
ClassificationTransformerLarge Language ModelVision Language ModelImageBiomedical Data
🎯 What it does: This paper proposes OpenPath, an open-set active learning framework for pathological image classification, addressing the cold start and OOD data selection issues.
Operating Room Workflow Analysis via Reasoning Segmentation over Digital Twins
Shen, Yiqing (Johns Hopkins University), Unberath, Mathias (University of Arkansas)
Object DetectionSegmentationDepth EstimationLarge Language ModelAgentic AIVideoMultimodality
🎯 What it does: A video inference segmentation framework for operating rooms (ORDiRS) based on digital twin (DT) representation and the corresponding ORDiRS-Agent is proposed for automated analysis of operating room workflows and efficiency.
Ophora: A Large-Scale Data-Driven Text-Guided Ophthalmic Surgical Video Generation Model
Li, Wei (Shanghai Jiao Tong University), He, Junjun (Shanghai Artificial Intelligence Laboratory)
GenerationData SynthesisSafty and PrivacyTransformerLarge Language ModelSupervised Fine-TuningVision Language ModelVideoText
🎯 What it does: This work presents Ophora, a text instruction-based model for generating ophthalmic surgery videos. It first constructs 160K high-quality video-instruction pairs through a complete data cleaning process and then progressively fine-tunes a pre-trained T2V model using this data, ultimately achieving real-time generation of realistic and reliable surgical videos based on doctor descriptions.
Opportunistic Osteoporosis Diagnosis via Texture-Preserving Self-Supervision, Mixture of Experts and Multi-Task Integration
Huang, Jiaxing (Chinese Academy of Sciences), Luo, Wei (Central South University)
ClassificationObject DetectionSegmentationConvolutional Neural NetworkTransformerMixture of ExpertsImageBiomedical DataComputed Tomography
🎯 What it does: Using the obtained chest and abdominal CT and DXA data, a unified deep learning framework was constructed to achieve CT-based opportunistic osteoporosis diagnosis, bone mineral density regression, and vertebral localization.
OralSAM: One-shot Segmentation for Intraoral Ultrasound Videos with Adaptive Feature Correlation and Self-prompting Strategy
Kumaralingam, Logiraj (University of Alberta), Le, Lawrence H. (University of Alberta)
SegmentationTransformerOptical FlowVideoBiomedical DataUltrasound
🎯 What it does: This paper proposes a one-shot video segmentation network named OralSAM, which can achieve segmentation of periodontal ultrasound videos using single-frame annotations.
OsteoOpt: A Bayesian Optimization Framework for Enhancing Bone Union Likelihood in Mandibular Reconstruction Surgery
Aftabi, Hamidreza (University of British Columbia), Fels, Sidney (University of British Columbia)
OptimizationBiomedical DataMagnetic Resonance ImagingComputed Tomography
🎯 What it does: A framework for bone healing prediction and surgical planning based on Bayesian optimization is proposed and implemented to improve the bone healing rate in mandibular reconstruction surgery.
OTSurv: A Novel Multiple Instance Learning Framework for Survival Prediction with Heterogeneity-aware Optimal Transport
Ren, Qin (Stony Brook University), You, Chenyu (Stony Brook University)
OptimizationExplainability and InterpretabilityBiomedical Data
🎯 What it does: A framework called OTSurv based on multi-instance learning is proposed, which models the global long-tail heterogeneity and local predictive uncertainty of WSI using optimal transport to achieve survival time prediction.
Out-of-Distribution Nuclei Segmentation in Histology Imaging via Liquid Neural Networks with Modern Hopfield Layer
Swain, Bishal Ranjan (Kumoh National Institute of Technology), Ko, Jaepil (Kumoh National Institute of Technology)
SegmentationDomain AdaptationImageBiomedical Data
🎯 What it does: A framework that combines liquid neural networks with modern Hopfield networks is proposed for handling discrete distribution kernel segmentation of omics images.
P2INR-FWI: an Implicit Neural Representation Method for Speed of Sound Image Reconstruction in Ultrasound Computed Tomography
Wang, Zesong (Huazhong University of Science and Technology), Qiu, Wu (Huazhong University of Science and Technology)
Biomedical DataComputed TomographyUltrasound
🎯 What it does: A USCT velocity image reconstruction method based on Polar Coordinate Implicit Neural Representation (P2 INR-FWI) is proposed, overcoming the sensitivity of traditional FWI to initial models and the problem of cycle skipping;
Paired Image Generation with Diffusion-Guided Diffusion Models
Zhang, Haoxuan (University of Science and Technology of China), Zheng, Jian (University of Science and Technology of China)
SegmentationGenerationData SynthesisDiffusion modelImageBiomedical DataMagnetic Resonance Imaging
🎯 What it does: This paper proposes a paired image generation method called PIG without external conditions, which simultaneously generates DBT slices and tumor masks in medical imaging, thereby providing complete annotated data.
Pairwise-Constrained Implicit Functions for 3D Human Heart Modeling
Le, Hieu (EPFL), Fua, Pascal (Stony Brook University)
SegmentationGenerationAuto EncoderImageBiomedical DataMagnetic Resonance ImagingComputed Tomography
🎯 What it does: A paired constraint implicit function method is proposed to construct a three-dimensional model of the multi-layer structure of the human heart, ensuring accurate and seamless shared surfaces between different components.
Parameter-Efficient 12-Lead ECG Reconstruction from a Single Lead
Lee, Junseok (Korea University), Yoo, Chuck (Korea University Anam Hospital)
RestorationComputational EfficiencyConvolutional Neural NetworkTime SeriesBiomedical DataElectrocardiogram
🎯 What it does: A parameter-efficient mEcgNet model is proposed to reconstruct a 12-lead electrocardiogram (ECG) from a single-lead I signal, suitable for wearable IoT devices.
Parameter-Efficient Fine-Tuning of 3D DDPM for MRI Image Generation Using Tensor Networks
Li, Binghua (Tokyo University of Agriculture and Technology), Sun, Zhe (Juntendo University)
GenerationData SynthesisConvolutional Neural NetworkDiffusion modelImageBiomedical DataMagnetic Resonance ImagingAlzheimer's Disease
🎯 What it does: A parameter-efficient fine-tuning method based on tensor networks, TenVOO, is proposed for the 3D U-Net structure of DDPM to generate brain MRI images.
Parameterized Diffusion Optimization enabled Autoregressive Ordinal Regression for Diabetic Retinopathy Grading
Yu, Qinkai (University of Exeter), Meng, Yanda (University Of Exeter)
ClassificationOptimizationTransformerDiffusion modelImage
🎯 What it does: A self-regressive sequence model AOR-DR is proposed for the grading task of diabetic retinopathy.
PATE: Enhancing Few-Shot Pathological Image Classification via Prompt-Based Text-Image Embedding Adaptation
Chen, Shenghao (University of Science and Technology of China), Zhou, S. Kevin (Hohai University)
ClassificationTransformerPrompt EngineeringImageMultimodality
🎯 What it does: The PATE framework is proposed, utilizing bidirectional deep prompts from both visual and textual modalities, along with a bridging function, to enhance the performance of few-shot pathological image classification.
Path Signature Features Revealed SSRI-Induced White Matter Morphological Reorganization in Depressions
Qin, Jiaolong, Wu, Ye (Nanjing University of Science and Technology)
Drug DiscoveryBiomedical DataMagnetic Resonance ImagingDiffusion Tensor Imaging
🎯 What it does: Using path signature (PS) features to perform macrostructural reconstruction and change detection of white matter in clinical-grade diffusion MRI data of patients with depression before and after SSRI treatment, revealing the reorganization of tracts such as TPT, ALIC, and SCC.
PathoCellBench: A Comprehensive Benchmark for Cell Phenotyping
Lüscher, Jérôme (Helmholtz Imaging), Rumberger, Josef Lorenz (Humboldt-Universität zu Berlin)
ClassificationSegmentationTransformerImageBiomedical DataBenchmark
🎯 What it does: Proposed the PathoCellBench benchmark, released the PathoCell dataset with 14 categories of cells, and systematically evaluated the performance of various pathological baseline models and supervised baselines in cell phenotype prediction.
Pathology Image Compression with Pre-trained Autoencoders
Yellapragada, Srikar (Stony Brook University), Samaras, Dimitris (Stony Brook University)
SegmentationCompressionDiffusion modelAuto EncoderImage
🎯 What it does: This paper utilizes pre-trained latent diffusion models (Stable Diffusion 1.5/3 and DC-AE) autoencoders for pathological image compression, and fine-tunes them by adding pathology-specific perceptual loss on the decoder, further employing K-means quantization to compress the latent variables.
Pathology Report Generation and Multimodal Representation Learning for Cutaneous Melanocytic Lesions
Lucassen, Ruben T. (University Medical Center Utrecht), Veta, Mitko (Eindhoven University of Technology)
GenerationRetrievalRepresentation LearningTransformerLarge Language ModelVision Language ModelContrastive LearningImageTextMultimodalityBiomedical Data
🎯 What it does: A visual-language model specifically designed for skin melanocytic lesions was trained and evaluated, capable of generating complete pathology reports based on whole-slide H&E images and achieving cross-modal retrieval.
Pathology-Aware Adaptive Watermarking for Text-Driven Medical Image Synthesis
Kim, Chanyoung (Yonsei University), Hwang, Seong Jae (Mediwhale)
GenerationData SynthesisDiffusion modelImageBiomedical DataMagnetic Resonance ImagingComputed Tomography
🎯 What it does: Designed and implemented the MedSign framework, embedding watermarks in text-driven medical image generation while preserving pathological information; dynamically adjusting watermark strength by locating pathological regions through cross-attention.
Pathology-aware Virtual H&E Staining of Section-free Thick Tissues with Semantic Contrastive Guidance
Oh, Jintaek (Hong Kong University of Science and Technology), Wong, Terence T. W. (Hong Kong University of Science and Technology)
Image TranslationGenerationGenerative Adversarial NetworkContrastive LearningImage
🎯 What it does: A virtual H&E staining framework based on semantic contrast guidance (SemCG-Stain) is proposed, which can directly perform optical imaging on thick tissues and generate equivalent H&E images.
Pathology-Informed Latent Diffusion Model for Anomaly Detection in Lymph Node Metastasis
Wang, Jiamu (Korea University), Kwak, Jin Tae (Korea University)
Anomaly DetectionVision Language ModelDiffusion modelImageBiomedical Data
🎯 What it does: A latent diffusion model based on pathological text prompts (AnoPILaD) is proposed for unsupervised detection of lymph node metastatic abnormalities.
PathoPainter: Augmenting Histopathology Segmentation via Tumor-aware Inpainting
Liu, Hong (Eindhoven University of Technology), Veta, Mitko (Eindhoven University of Technology)
RestorationSegmentationData SynthesisTransformerDiffusion modelAuto EncoderImage
🎯 What it does: The PathoPainter framework is proposed, which generates high-quality, aligned digital pathology image-mask pairs through an image inpainting method based on tumor masks and regional tumor embeddings.
PathoPrompt: Cross-Granular Semantic Alignment for Medical Pathology Vision-Language Models
Huang, Runlin (Beijing Normal-Hong Kong Baptist University), Su, Weifeng (Beijing Normal-Hong Kong Baptist University)
ClassificationKnowledge DistillationTransformerVision Language ModelImageBiomedical Data
🎯 What it does: The PathoPrompt framework is proposed to enhance the fine-grained performance of pathological image classification through multi-granularity semantic alignment.
PathVG: A New Benchmark and Dataset for Pathology Visual Grounding
Zhong, Chunlin (Huazhong University of Science and Technology), Bai, Xiang (Huazhong University of Science and Technology)
Object DetectionSegmentationConvolutional Neural NetworkTransformerLarge Language ModelSupervised Fine-TuningVision Language ModelImageTextBiomedical DataBenchmark
🎯 What it does: This paper proposes the PathVG pathological visual localization benchmark and constructs the RefPath dataset. It then designs the PKNet model to enhance pathological text understanding through LLM, completing text-based localization of pathological image regions.
Patient-specific radiomic feature selection with reconstructed healthy persona of knee MR images
Chen, Yaxi (University College London), Hu, Yipeng (King's College London)
ClassificationAnomaly DetectionConvolutional Neural NetworkDiffusion modelImageBiomedical DataMagnetic Resonance Imaging
🎯 What it does: This paper proposes a patient-specific feature selection framework that combines generative models and radiomics for disease diagnosis in knee MRI.
Pattern-Anchored Adaptive Prototype Learning for Gastroscopic Lesion Detection and Beyond
Zhang, Xuanye (Shenzhen Research Institute of Big Data), Xiong, Yuanhuan (Jiangxi Provincial Maternal and Child Health Hospital)
Object DetectionExplainability and InterpretabilityConvolutional Neural NetworkContrastive LearningImageBiomedical Data
🎯 What it does: A Pattern-Anchored Adaptive Prototype Learning (PAAPL) framework is proposed, which implements subcategory pattern detection of endoscopic lesions through a prototype branch and enhances interpretability and detection performance under limited annotations through pattern-anchored adaptive learning.
PCR-MIL: Phenotype Clustering Reinforced Multiple Instance Learning for Whole Slide Image Classification
Lou, Jingjiao (Shanghai Jiao Tong University), Ji, Bing (Shanghai Jiao Tong University)
ClassificationOptimizationConvolutional Neural NetworkReinforcement LearningImage
🎯 What it does: This paper proposes a pseudo-bag-based multi-instance learning framework that utilizes phenotypic clustering for feature selection and an enhanced feature extractor to reduce noisy labels in pathological slices and improve feature discriminability, ultimately achieving more accurate diagnostic inference.
PD-INR: Prior-Driven Implicit Neural Representations for TOF-PET Reconstruction
Long, Yuxuan (Zhejiang Lab), Zhu, Wentao (Westlake University)
RestorationOptimizationNeural Radiance FieldBiomedical DataPositron Emission Tomography
🎯 What it does: A self-supervised TOF-PET reconstruction framework is proposed, utilizing implicit neural representation (INR) to model PET images, and achieving reconstruction through differential forward projection and beam-level TV regularization.
PD-UniST: Prompt-Driven Universal Model for Unpaired H&E-to-IHC Stain Translation
Zhang, Chujie (Ritsumeikan University), Chen, Yen-Wei (Ritsumeikan University)
Image TranslationGenerationPrompt EngineeringGenerative Adversarial NetworkContrastive LearningImageBiomedical Data
🎯 What it does: A Prompt-Driven Universal Model (PD-UniST) is proposed, capable of converting H&E stained images into various IHC stained images without paired training, supporting the generation of staining results for different tumor markers with a single training session.
PDC-Net: Pattern Divide-and-Conquer Network for Pelvic Radiation Injury Segmentation
Xiong, Xinyu (Sun Yat-sen University), Qin, Qiyuan (Sun Yat-sen University)
SegmentationConvolutional Neural NetworkMixture of ExpertsImageBiomedical DataMagnetic Resonance Imaging
🎯 What it does: This paper proposes the PDC-Net network for the automatic segmentation of pelvic radiotherapy injury (PRI) regions from MRI images.
PDF-Net: Prototype-Aware Dynamic Fusion Network for Nasopharyngeal Carcinoma T-staging Classification with Epstein-Barr Virus DNA
Lu, Wantong (Southern Medical University), Lu, Lijun (Southern Medical University)
ClassificationTransformerMultimodalityBiomedical DataMagnetic Resonance Imaging
🎯 What it does: This paper proposes a multi-modal Prototype-Aware Dynamic Fusion Network (PDF-Net) that combines MR imaging and EBV DNA data for NPC T staging classification.
PedCLIP: A Vision-Language model for Pediatric X-rays with Mixture of Body part Experts
Huy, Ta Duc (Australian Institute for Machine Learning), Phan, Minh Hieu (Australian Institute for Machine Learning)
ClassificationSegmentationTransformerMixture of ExpertsVision Language ModelContrastive LearningImageMultimodality
🎯 What it does: We propose PedCLIP, a vision-language pre-training model specifically designed for pediatric X-rays.
PerioDet: Large-Scale Panoramic Radiograph Benchmark for Clinical-Oriented Apical Periodontitis Detection
Fang, Xiaocheng (Beijing Institute of Technology), Chen, Bingzhi (Shenzhen University)
Object DetectionConvolutional Neural NetworkImageBenchmark
🎯 What it does: This paper presents a large-scale panoramic dental X-ray dataset, PerioXrays, and proposes a clinically-oriented framework for detecting apical periodontitis, PerioDet, which combines a background denoising attention module (BDA) and an IoU dynamic calibration module (IDC) to achieve automatic detection of apical periodontitis.
Personalized Federated Side-Tuning for Medical Image Classification
Chen, Jiayi (Northwestern Polytechnical University), Xia, Yong (Northwestern Polytechnical University)
ClassificationFederated LearningTransformerVision Language ModelContrastive LearningImageBiomedical DataMagnetic Resonance Imaging
🎯 What it does: This paper proposes and implements a personalized federated side-tuning framework, pFedST, for multi-center medical image classification, which can effectively adapt to the data heterogeneity of each center without sharing the original data.
PFESA: FFT-based Parameter-Free Edge and Structure Attention for Medical Image Segmentation
Li, Mingqian (South China Normal University), Ma, Qiongxiong (South China Normal University)
SegmentationConvolutional Neural NetworkTransformerImageBiomedical DataMagnetic Resonance ImagingComputed Tomography
🎯 What it does: This paper proposes a parameter-free edge and structure attention mechanism based on FFT (PFESA), which enhances edge and structure preservation in medical image segmentation by separating high and low-frequency features in the frequency domain and generating dual attention based on signal-to-noise ratio.
Phase-Informed Tool Segmentation for Manual Small-Incision Cataract Surgery
Sachdeva, Bhuvan (Microsoft Research), Jain, Mohit (Microsoft Research)
SegmentationConvolutional Neural NetworkImageVideo
🎯 What it does: A tool segmentation framework called ToolSeg is proposed, which is oriented towards surgical phase information, and the first pixel-level annotated dataset Sankara-MSICS is provided based on MSICS surgery.
Phenotype Representation and Analysis via Discriminative Atypicality (PRADA) to capture the structural heterogeneity of Autism Spectrum Disorder
Onemli, Emre (University of North Carolina at Chapel Hill), Styner, Martin (University of North Carolina at Chapel Hill)
Anomaly DetectionImageBiomedical DataMagnetic Resonance Imaging
🎯 What it does: A framework called PRADA is proposed, which combines Multi-Scale Score Matching Analysis (MSMA) with Growing Hierarchical Self-Organizing Maps (GHSOM) to identify subtypes of Autism Spectrum Disorder (ASD) in structural MRI data and associate brain structural abnormalities with behavioral/cognitive measures.
Phenotype-Guided Generative Model for High-Fidelity Cardiac MRI Synthesis: Advancing Pretraining and Clinical Applications
Li, Ziyu (Beijing Institute of Technology), Huang, Zhengxing (First Affiliated Hospital of Zhejiang University School of Medicine)
GenerationData SynthesisTransformerDiffusion modelAuto EncoderVideoBiomedical DataMagnetic Resonance Imaging
🎯 What it does: A two-stage cardiac MRI synthesis framework (CPGG) is proposed, which first generates the cardiac phenotype distribution and then generates high-fidelity cardiac CINE sequences conditionally based on the phenotype using a masked autoregressive diffusion model.
Phrase-grounded Fact-checking for Automatically Generated Chest X-ray Reports
Mahmood, Razi (Rensselaer Polytechnic Institute), Syeda-Mahmood, Tanveer (Rensselaer Polytechnic Institute)
ClassificationObject DetectionExplainability and InterpretabilityTransformerVision Language ModelContrastive LearningImageMultimodalityBiomedical Data
🎯 What it does: This paper proposes a fact-checking model based on phrase localization (FC model) for automatically identifying findings and localization errors in chest X-ray reports during the clinical reasoning phase. By perturbing the findings and locations of real reports, 27 million synthetic samples were generated to train a multi-label cross-modal contrastive regression network, which predicts the authenticity and corresponding image location for each finding during inference, achieving error detection and interpretable localization for generated reports.
Physics informed guided diffusion for accelerated multi-parametric MRI reconstruction
Mayo, Perla (University of Bristol), Golbabaee, Mohammad (University of Bristol)
Diffusion modelBiomedical DataMagnetic Resonance Imaging
🎯 What it does: Proposes MRF-DiPh, a physics-informed diffusion model for accelerating multi-parameter MRI reconstruction.
Physics-driven Signal Regularization in Diffusion Models for Multi-contrast MR Image Synthesis
Shin, Yejee (Yonsei University), Kim, Sewon (NAVER Cloud)
GenerationData SynthesisDiffusion modelImageBiomedical DataMagnetic Resonance ImagingPhysics Related
🎯 What it does: A multi-contrast MR image synthesis framework based on diffusion models, MRDiff, has been designed to synthesize target contrast images from existing multi-contrast MR images.
Physics-guided deep image prior network for general zero-shot stain deconvolution
Chen, Jianan (University of Toronto), Martel, Anne L. (University of Toronto)
RestorationConvolutional Neural NetworkImageBiomedical DataPhysics Related
🎯 What it does: A zero-shot, physics-guided deep image prior network (PGDIPS) is proposed for general staining deconvolution and normalization.
Physics-Informed Implicit Neural Representations for Joint B0 Estimation and Echo Planar Imaging
Huang, Wenqi (Technical University of Munich), Setsompop, Kawin (Stanford University)
RestorationImageBiomedical DataMagnetic Resonance Imaging
🎯 What it does: A physics-informed implicit neural representation (PINR) framework is proposed for the joint estimation of the B0 field and the reconstruction of distortion-free EPI images.
Physics-Informed Neural ODEs for Temporal Dynamics Modeling in Cardiac T1 Mapping
Capitão, Nuno (Delft University of Technology), Tao, Qian (Delft University of Technology)
Recurrent Neural NetworkBiomedical DataMagnetic Resonance ImagingPhysics RelatedOrdinary Differential Equation
🎯 What it does: A framework for accelerated cardiac T1 mapping based on physics-informed neural ODEs is proposed, achieving high-precision T1 estimation under sparse sampling.
Physics-Informed Neural Operators for Tissue Elasticity Reconstruction
Kim, Youjin (Chung-Ang University), Kwon, Junseok (Chung-Ang University)
Biomedical DataMagnetic Resonance ImagingPhysics Related
🎯 What it does: This paper proposes a magnetic resonance elastography (MRE) elastic reconstruction framework based on a physics-informed neural operator (MRE-Hyper), which can learn the mapping from wave images to elastic fields in one go;
Physiological neural representation for personalised tracer kinetic parameter estimation from dynamic PET
Tehlan, Kartikay (University Hospital Augsburg), Wendler, Thomas (University Hospital Augsburg)
Biomedical DataComputed TomographyPositron Emission TomographyOrdinary Differential Equation
🎯 What it does: Using implicit neural representations (INR) for voxel-level reversible two-compartment kinetic model (TCKM) parameter estimation of dynamic [18F]FDG PET data, constructing continuous spatiotemporal parameter mappings;
PLUS: Plug-and-Play Enhanced Liver Lesion Diagnosis Model on Non-Contrast CT Scans
Hao, Jiacheng (Tsinghua University), Yan, Ke (China Medical University)
SegmentationConvolutional Neural NetworkGraph Neural NetworkImageBiomedical DataComputed Tomography
🎯 What it does: A pluggable plugin framework called PLUS is proposed to enhance the existing 3D segmentation models for liver lesion diagnosis in non-contrast CT scans.