MICCAI 2024 Papers — Page 3
International Conference on Medical Image Computing and Computer-Assisted Intervention · 856 papers
Deep Learning for Cancer Prognosis Prediction Using Portrait Photos by StyleGAN Embedding
Hagag, Amr (Friedrich-Alexander-Universität Erlangen-Nürnberg), Putz, Florian (Friedrich-Alexander-Universität Erlangen-Nürnberg)
ClassificationGenerationExplainability and InterpretabilityConvolutional Neural NetworkGenerative Adversarial NetworkImage
🎯 What it does: Using latent space embeddings generated by StyleGAN2, facial photos of cancer patients are combined with clinical features to predict overall survival using the CoxPH and DeepSurv models, and 'health attributes' are constructed through linear operations in the latent space to achieve interpretable image editing.
Deep Model Reference: Simple yet Effective Confidence Estimation for Image Classification
Zheng, Yuanhang (Sun Yat-sen University), Wang, Ruixuan (Sun Yat-sen Univerisity)
ClassificationAnomaly DetectionConvolutional Neural NetworkSupervised Fine-TuningImageBiomedical DataMagnetic Resonance ImagingComputed Tomography
🎯 What it does: This paper proposes a post-processing confidence estimation method based on Deep Model Reference (DMR) for misclassification detection in image classification.
Deep Spectral Methods for Unsupervised Ultrasound Image Interpretation
Tmenova, Oleksandra (Technical University of Munich), Navab, Nassir (Technische Universität München)
SegmentationTransformerContrastive LearningImageBiomedical DataUltrasound
🎯 What it does: A framework for unsupervised ultrasound image segmentation based on self-supervised Transformer features and deep spectral clustering is proposed.
Deep Volume Reconstruction from Multi-focus Microscopic Images
Azevedo, Caio (École Polytechnique), Morooka, Ken'ichi (Kumamoto University)
RestorationOptimizationConvolutional Neural NetworkImage
🎯 What it does: Reconstruct the three-dimensional volume of transparent samples such as cells using multi-focal microscopy images through the Deep Image Prior (DIP) framework.
Deep-learning-based groupwise registration for motion correction of cardiac T1 mapping
Zhang, Yi (Sichuan University), Tao, Qian (Delft University of Technology)
Convolutional Neural NetworkBiomedical DataMagnetic Resonance Imaging
🎯 What it does: This paper proposes a template-free deep learning group registration framework for motion correction in cardiac T1 mapping.
DeepRepViz: Identifying potential confounders in deep learning model predictions
Rane, Roshan Prakash (Charité - Universitätsmedizin Berlin), Ritter, Kerstin (Charité - Universitätsmedizin Berlin)
Explainability and InterpretabilityConvolutional Neural NetworkImageBiomedical DataMagnetic Resonance Imaging
🎯 What it does: A framework named DeepRepViz is proposed to detect potential confounding factors (such as gender, age, etc.) in deep learning models for neuroimaging prediction tasks and to enhance the interpretability of the models.
Deform-Mamba Network for MRI Super-Resolution
Ji, Zexin (Central South University), Ruan, Su (University of Rouen-Normandy)
RestorationSuper ResolutionConvolutional Neural NetworkImageBiomedical DataMagnetic Resonance Imaging
🎯 What it does: A network named Deform-Mamba is proposed for upsampling low-resolution MRI images to high resolution.
Deform3DGS: Flexible Deformation for Fast Surgical Scene Reconstruction with Gaussian Splatting
Yang, Shuojue (National University of Singapore), Jin, Yueming (University of Oxford)
Gaussian SplattingVideoPoint Cloud
🎯 What it does: The Deform3DGS framework is proposed, achieving rapid deformable tissue reconstruction by using 3D Gaussian Splatting and a flexible deformation model in surgical videos.
Deformation-Aware Segmentation Network Robust to Motion Artifacts for Brain Tissue Segmentation using Disentanglement Learning
Jung, Sunyoung (Yonsei University), Kim, Dong-Hyun (Yonsei University)
SegmentationConvolutional Neural NetworkSupervised Fine-TuningImageBiomedical DataMagnetic Resonance Imaging
🎯 What it does: An end-to-end network is proposed, integrating motion correction, decomposition learning, and brain tissue segmentation.
Depth-Aware Endoscopic Video Inpainting
Zhang, Francis Xiatian (University of Edinburgh), Shum, Hubert P. H. (Durham University)
RestorationDepth EstimationConvolutional Neural NetworkTransformerGenerative Adversarial NetworkVideo
🎯 What it does: This paper proposes a deep perception-based endoscopic video restoration framework called DAEVI, which achieves more realistic 3D spatial detail recovery by directly inferring depth from visual features, dual-modal paired channel fusion, and a depth-enhanced discriminator.
Depth-Driven Geometric Prompt Learning for Laparoscopic Liver Landmark Detection
Pei, Jialun (Chinese University of Hong Kong), Heng, Pheng-Ann (Chinese University of Hong Kong)
Object DetectionSegmentationConvolutional Neural NetworkContrastive LearningVideo
🎯 What it does: This paper proposes a deep-driven geometric prompt learning network for laparoscopic liver landmark detection, named D2GPLand, and releases a new L3D dataset.
DermaVQA: A Multilingual Visual Question Answering Dataset for Dermatology
Yim, Wen-wai (Microsoft Health AI), Xia, Fei (University of Washington)
TransformerLarge Language ModelVision Language ModelImageTextMultimodality
🎯 What it does: A multilingual (Chinese and English) dermatology visual question answering (Dermatology VQA) dataset called DermaVQA has been constructed, and baseline evaluations of three state-of-the-art multimodal large language models (Gemini-Pro Vision, GPT-4-vision-preview, LLaVA-FT+GPT-4) have been conducted on this dataset to explore the task of generating doctor responses.
DES-SAM: Distillation-Enhanced Semantic SAM for Cervical Nuclear Segmentation with Box Annotation
Huang, Lina (Central South University), Liu, Jianfeng (Central South University)
SegmentationKnowledge DistillationImage
🎯 What it does: A box supervision cervical nucleus segmentation network DES-SAM based on self-distillation prompts is proposed, aiming to reduce manual annotation costs and improve the model's generalization ability.
DeSAM: Decoupled Segment Anything Model for Generalizable Medical Image Segmentation
Gao, Yifan (University of Science and Technology of China), Gao, Xin (Chinese Academy of Sciences)
SegmentationDomain AdaptationTransformerImageBiomedical DataMagnetic Resonance ImagingComputed Tomography
🎯 What it does: This paper studies a model to enhance the generalization of medical image segmentation under single-source domain transfer—Decoupled Segment Anything Model (DeSAM), which achieves automated segmentation by decoupling the association between prompts and masks.
Design as Desired: Utilizing Visual Question Answering for Multimodal Pre-training
Su, Tongkun (Shenzhen Institute of Advanced Technology Chinese Academy of Sciences), Hu, Ying (Shenzhen Institute of Advanced Technology Chinese Academy of Sciences)
ClassificationSegmentationGenerationTransformerVision Language ModelContrastive LearningImageTextMultimodalityUltrasound
🎯 What it does: This paper proposes the use of Visual Question Answering (VQA) for pre-training on medical multimodal data and designs multi-granularity question-answer pairs to guide the model in learning pathological features.
Detecting noisy labels with repeated cross-validations
Chen, Jianan (University of Toronto), Martel, Anne L. (University of Toronto)
ClassificationData-Centric LearningTabular
🎯 What it does: Proposed and implemented a non-parametric noise detection algorithm based on repeated cross-validation, ReCoV, and its efficient variant, fastReCoV, for actively identifying erroneous labels in training data.
Development of Effective Connectome from Infancy to Adolescence
Li, Guoshi (University of North Carolina at Chapel Hill), Yap, Pew-Thian (University of North Carolina at Chapel Hill)
Time SeriesBiomedical DataMagnetic Resonance Imaging
🎯 What it does: Using regression DCM to model effective connectivity (EC) of rs-fMRI data from infants to adolescents aged 0-22 years, we mapped the nonlinear developmental trajectory of EC across the entire brain region with age and revealed age effects in different brain networks and specific brain areas.
Diff-VPS: Video Polyp Segmentation via a Multi-task Diffusion Network with Adversarial Temporal Reasoning
Lu, Yingling (Hong Kong University of Science and Technology), Zhu, Lei (Hong Kong University of Science and Technology)
Object DetectionSegmentationTransformerDiffusion modelGenerative Adversarial NetworkVideo
🎯 What it does: A multi-task video polyp segmentation framework based on diffusion models, Diff-VPS, is proposed, which enhances the segmentation capability for high-fidelity colored polyps through a temporal reasoning module.
Diff3Dformer: Leveraging Slice Sequence Diffusion for Enhanced 3D CT Classification with Transformer Networks
Jin, Zihao (Imperial College London), Yang, Guang (Sun Yat-sen University)
ClassificationTransformerDiffusion modelAuto EncoderImageBiomedical DataComputed Tomography
🎯 What it does: This paper proposes the Diff3Dformer model, which utilizes a diffusion autoencoder to extract semantic features from CT slices. It then generates slice prototypes through spherical K-means clustering and uses the cluster numbers as auxiliary input to a clustering attention Transformer to complete 3D CT classification.
DiffDGSS: Generalizable Retinal Image Segmentation with Deterministic Representation from Diffusion Models
Xie, Yingpeng (Shenzhen University), Lei, Baiying (Shenzhen University)
SegmentationDomain AdaptationDiffusion modelImage
🎯 What it does: This paper proposes the DiffDGSS framework, which achieves domain generalization for retinal image semantic segmentation by extracting deterministic latent representations from a pre-trained DDPM.
Differentiable Score-Based Likelihoods: Learning CT Motion Compensation From Clean Images
Thies, Mareike (Friedrich-Alexander-Universität Erlangen-Nürnberg), Maier, Andreas (Friedrich-Alexander-Universität Erlangen)
RestorationOptimizationDiffusion modelScore-based ModelImageBiomedical DataComputed TomographyOrdinary Differential Equation
🎯 What it does: This paper trains a score diffusion model solely on non-motion CT images to compute the exact likelihood of any CT image, using the likelihood gradient for gradient descent optimization of motion parameters, thereby achieving CT motion compensation.
Differentiable Soft Morphological Filters for Medical Image Segmentation
Guzzi, Lisa (Université Côte d'Azur), Delingette, Hervé (Université Côte d'Azur)
SegmentationConvolutional Neural NetworkImageBiomedical DataComputed Tomography
🎯 What it does: The study proposes a Microsoft morphological filter in medical image segmentation and integrates it into the network's loss function or final layer.
DiffExplainer: Unveiling Black Box Models Via Counterfactual Generation
Fang, Yingying, Yang, Guang (Sun Yat-sen University)
GenerationExplainability and InterpretabilityKnowledge DistillationDiffusion modelAuto EncoderImageBiomedical DataComputed Tomography
🎯 What it does: This paper proposes DiffExplainer, a counterfactual generation framework that combines diffusion autoencoders and teacher-student learning to reveal the decision basis of black-box models in medical CT image classification.
DiffRect: Latent Diffusion Label Rectification for Semi-supervised Medical Image Segmentation
Liu, Xinyu (Chinese University of Hong Kong), Yuan, Yixuan (Southern University of Science and Technology)
SegmentationConvolutional Neural NetworkDiffusion modelImageBiomedical DataMagnetic Resonance Imaging
🎯 What it does: This paper proposes a semi-supervised medical image segmentation framework named DiffRect, which combines pseudo-label correction with a latent diffusion model to enhance segmentation performance.
DiffuseReg: Denoising Diffusion Model for Obtaining Deformation Fields in Unsupervised Deformable Image Registration
Zhuo, Yongtai (Shanghai Jiao Tong University), Shen, Yiqing (Johns Hopkins University)
TransformerDiffusion modelImageBiomedical DataMagnetic Resonance Imaging
🎯 What it does: This paper proposes DiffuseReg, which achieves unsupervised deformable image registration by gradually denoising the displacement field using a diffusion model.
Diffusion as Sound Propagation: Physics-inspired Model for Ultrasound Image Generation
Domínguez, Marina (Technical University of Munich), Azampour, Mohammad Farid (Technical University of Munich)
GenerationData SynthesisDiffusion modelImageBiomedical DataUltrasoundPhysics Related
🎯 What it does: The study introduces a physics-inspired B-Maps noise scheduler in ultrasound image generation to improve diffusion models, better simulating sound wave attenuation and achieving more realistic synthetic B-mode ultrasound images.
Diffusion Models with Implicit Guidance for Medical Anomaly Detection
Bercea, Cosmin I. (Technical University of Munich), Schnabel, Julia A. (Technical University of Munich)
RestorationSegmentationAnomaly DetectionDiffusion modelImageBiomedical DataMagnetic Resonance Imaging
🎯 What it does: The THOR (Temporal Harmonization for Optimal Restoration) method is proposed and implemented, which introduces intermediate masks during the reverse diffusion process for implicit guidance in unsupervised medical anomaly detection, enhancing the accuracy of image restoration and achieving finer anomaly segmentation.
Diffusion-based Domain Adaptation for Medical Image Segmentation using Stochastic Step Alignment
Ji, Wen (Hong Kong University of Science and Technology), Chung, Albert C. S. (Hong Kong University of Science and Technology)
SegmentationDomain AdaptationDiffusion modelImageBiomedical DataMagnetic Resonance ImagingComputed Tomography
🎯 What it does: This paper proposes an unsupervised domain adaptation framework based on a diffusion model, which obtains intermediate images during the multi-step generation process from the source domain to the target domain, and achieves feature and prediction space alignment through random step domain alignment, thereby enhancing cross-modal medical image segmentation performance.
Diffusion-based Generative Image Outpainting for Recovery of FOV-Truncated CT Images
Liman, Michelle Espranita (Technical University of Munich), Müller, Philip (Technical University of Munich)
RestorationGenerationDiffusion modelImageComputed Tomography
🎯 What it does: To address the limitations of body composition analysis caused by field of view (FOV) truncation in chest CT scans, this paper proposes an image outpainting method based on diffusion models to automatically restore truncated slices, thereby enabling complete measurements of muscle and fat areas.
Diffusion-Enhanced Transformation Consistency Learning for Retinal Image Segmentation
Li, Xiang (Nanyang Technological University), Duan, Lixin (University of Electronic Science and Technology of China)
SegmentationDiffusion modelImage
🎯 What it does: A semi-supervised retinal image segmentation framework named DiffTCL is proposed, which enhances segmentation accuracy under limited annotations through self-supervised diffusion pre-training and transformation consistency learning.
DINO-Reg: General Purpose Image Encoder for Training-free Multi-modal Deformable Medical Image Registration
Song, Xinrui (Rensselaer Polytechnic Institute), Yan, Pingkun (Rensselaer Polytechnic Institute)
SegmentationOptimizationTransformerContrastive LearningImageMultimodalityBiomedical DataMagnetic Resonance ImagingComputed Tomography
🎯 What it does: A training-independent DINO-Reg method is proposed, which utilizes the pre-trained DINOv2 visual backbone model to extract semantic features and achieves multi-modal deformable medical image registration through rigid slice matching and gradient descent refinement.
DinoBloom: A Foundation Model for Generalizable Cell Embeddings in Hematology
Koch, Valentin (Technical University of Munich), Marr, Carsten (Helmholtz Zentrum München - German Research Center for Environmental Health)
ClassificationRepresentation LearningTransformerContrastive LearningImageBiomedical Data
🎯 What it does: DinoBloom has been constructed and made public, a large-scale self-supervised pre-training model specifically designed for hematology single-cell images, trained using 13 publicly available datasets, achieving excellent generalization performance in tasks such as cell type classification and AML subtype classification.
DiRecT: Diagnosis and Reconstruction Transformer for Mandibular Deformity Assessment
Xu, Xuanang (Rensselaer Polytechnic Institute), Yan, Pingkun (Rensselaer Polytechnic Institute)
TransformerImageComputed Tomography
🎯 What it does: Using CBCT/3dMD images, 328 3D facial soft tissue landmark points were extracted through the MediaPipe 2D landmark model, and then these landmark points were diagnosed using the Diagnosis-Reconstruction Transformer (DiRecT), supplemented by a reconstruction task and teacher-student semi-supervised learning to enhance model performance.
Disease Progression Prediction Incorporating Genotype-Environment Interactions: A Longitudinal Neurodegenerative Disorder Study
Zhang, Jin (Northwestern Polytechnical University), Du, Lei (Northwestern Polytechnical University)
Explainability and InterpretabilityComputational EfficiencyBiomedical DataMagnetic Resonance ImagingAlzheimer's Disease
🎯 What it does: This paper proposes an interpretable mutual learning framework MA‑DPxGE for predicting the progression of Alzheimer's disease and identifying biomarkers based on longitudinal brain imaging and gene, protein, and environmental interactions.
Disease-informed Adaptation of Vision-Language Models
Zhang, Jiajin (Rensselaer Polytechnic Institute), Yan, Pingkun (Rensselaer Polytechnic Institute)
ClassificationSegmentationTransformerPrompt EngineeringVision Language ModelImageMultimodalityBiomedical Data
🎯 What it does: Using a pre-trained vision-language model, this paper addresses the diagnostic challenges of underrepresented or newly emerging diseases with few samples through disease knowledge-driven prompts and prototype learning.
Disentangled Attention Graph Neural Network for Alzheimer’s Disease Diagnosis
Gamgam, Gurur (Bogazici University), Acar, Burak (Koc University)
ClassificationExplainability and InterpretabilityGraph Neural NetworkGraphBiomedical DataAlzheimer's Disease
🎯 What it does: The Disentangled Attention Graph Neural Network (DAGNN) model is proposed for the diagnosis of Alzheimer's Disease (ADD) and Mild Cognitive Impairment (MCI) based on structural networks.
Disentangled Hybrid Transformer for Identification of Infants with Prenatal Drug Exposure
Cheng, Jiale (University of North Carolina at Chapel Hill), Li, Gang (University of North Carolina at Chapel Hill)
ClassificationRecognitionTransformerImageBiomedical DataMagnetic Resonance Imaging
🎯 What it does: The study uses a hybrid volume-surface Transformer model to identify prenatal drug exposure in infants.
Distributionally-Adaptive Variational Meta Learning for Brain Graph Classification
Du, Jing (Macquarie University), Giral, Alexis (Systemethix)
ClassificationMeta LearningGraph Neural NetworkAuto EncoderGraphBiomedical Data
🎯 What it does: A distribution-adaptive variational meta-learning framework (DAML) is proposed for brain image classification, combining graph representation learning and distribution-adaptive variational meta-learning to address data drift and small sample issues.
Diversified and Structure-realistic Fundus Image Synthesis for Diabetic Retinopathy Lesion Segmentation
Feng, Xiaoyi (Chinese University of Hong Kong), Yuan, Wu (Chinese University of Hong Kong)
SegmentationGenerationData SynthesisConvolutional Neural NetworkTransformerDiffusion modelImage
🎯 What it does: A two-stage diffusion model is proposed (Stage I generates structure-guided DR lesion masks; Stage II performs lesion repair based on the masks in healthy fundus images) to generate diverse and anatomically realistic fundus images and corresponding annotations under limited labeled data, used for data augmentation in segmentation models.
DnFPlane For Efficient and High-Quality 4D Reconstruction of Deformable Tissues
Bu, Ran (China University of Mining and Technology), Wang, Hesheng (Shanghai Jiaotong University)
RestorationData SynthesisDepth EstimationRecurrent Neural NetworkNeural Radiance FieldVideoBiomedical DataMagnetic Resonance Imaging
🎯 What it does: A method named DnFPlane is proposed to achieve efficient and high-definition 4D reconstruction of deformable tissues, particularly addressing the issues of depth distortion and dynamic blur caused by occlusion from surgical instruments.
Domain Adaptation for Unsupervised Cancer Detection: An application for skin Whole Slides Images from an interhospital dataset
P. García-de-la-Puente, Natalia (Universitat Politècnica de València), Naranjo, Valery (Universitat Politècnica de València)
Domain AdaptationAnomaly DetectionAuto EncoderImage
🎯 What it does: This paper proposes a domain-adaptive unsupervised cancer detection method (DAUD) that utilizes autoencoders and latent variables to identify malignant and unknown cases in skin tissue slices.
Domain Adaptation of Echocardiography Segmentation Via Reinforcement Learning
Judge, Arnaud (University of Sherbrooke), Jodoin, Pierre-Marc (Universit' de Sherbrooke)
SegmentationDomain AdaptationConvolutional Neural NetworkReinforcement LearningImageBiomedical DataUltrasound
🎯 What it does: Developed the RL4Seg framework, utilizing reinforcement learning to achieve unsupervised 2D ultrasound cardiac image segmentation and uncertainty estimation, significantly reducing the need for expert annotations.
DomainAdapt: Leveraging Multitask Learning and Domain Insights for Children’s Nutritional Status Assessment
Khan, Misaal (Indian Institute of Technology Jodhpur), Singh, Kuldeep (All India Institute of Medical Science Jodhpur)
ClassificationDomain AdaptationConvolutional Neural NetworkImage
🎯 What it does: A DomainAdapt multi-task learning framework was constructed to achieve automatic assessment of nutritional status using children's multi-pose images, and the AnthroVision large-scale multi-angle image dataset of children was released for the first time.
Double-tier Attention based Multi-label Learning Network for Predicting Biomarkers from Whole Slide Images of Breast Cancer
Wang, Mingkang (Dalian University of Technology), Xu, Hongming (Dalian University Of Technology)
ClassificationConvolutional Neural NetworkTransformerImageBiomedical Data
🎯 What it does: This paper proposes a Dual-layer Attention Multi-label Learning Network (DAMLN) that can simultaneously predict the expression status of estrogen receptor (ER), progesterone receptor (PR), and human epidermal growth factor receptor 2 (HER2) using only H&E stained whole slide images (WSI) of breast cancer.
DPMNet: Dual-Path MLP-based Network for Aneurysm Image Segmentation
Wang, Shudong (China University of Petroleum), Qiao, Sibo (Tiangong University)
SegmentationConvolutional Neural NetworkImageBiomedical DataMagnetic Resonance Imaging
🎯 What it does: DPMNet is proposed—a dual-path MLP network for intracranial aneurysm image segmentation, which combines global and local branches, CNN and MLP modules to achieve multi-scale feature fusion.
DRIM: Learning Disentangled Representations from Incomplete Multimodal Healthcare Data
Robinet, Lucas (IUCT-Oncopole-Institut Claudius Regaud), Cohen-Jonathan Moyal, Elizabeth (IUCT-Oncopole-Institut Claudius Regaud)
OptimizationRepresentation LearningTransformerContrastive LearningMultimodalityBiomedical DataMagnetic Resonance ImagingElectronic Health Records
🎯 What it does: Proposes the DRIM framework, which decomposes incomplete multimodal medical data using shared and unique encoders, and further integrates them to achieve survival prediction.
DSCENet: Dynamic Screening and Clinical-Enhanced Multimodal Fusion for MPNs Subtype Classification
Zhang, Yuan (Southeast University), Yang, Guanyu (Southeast University)
ClassificationConvolutional Neural NetworkMultimodalityBiomedical Data
🎯 What it does: A multi-modal fusion network based on dynamic selection and clinical enhancement (DSCENet) is proposed for the precise classification of subtypes of myeloproliferative neoplasms (MPN).
DSNet: A Spatio-Temporal Consistency Network for Cerebrovascular Segmentation in Digital Subtraction Angiography Sequences
Xie, Qihang (Ningbo Institute of Materials Technology and Engineering, Chinese Academy of Sciences), Zhang, Jiong (Chinese Academy of Sciences)
SegmentationConvolutional Neural NetworkImageVideo
🎯 What it does: This paper proposes a dual-branch spatiotemporal consistency network, DSNet, which simultaneously extracts dynamic blood flow and static vascular structures using DSA sequences and their MIP images, achieving precise segmentation of cerebral vessels.
DTCA: Dual-Branch Transformer with Cross-Attention for EEG and Eye Movement Data Fusion
Zhang, Xiaoshan (Northwestern Polytechnical University), Zhang, Shu (Northwestern Polytechnical University)
ClassificationRecognitionTransformerMultimodalityBiomedical Data
🎯 What it does: This paper proposes a dual-branch Transformer-based cross-attention model (DTCA) for the fusion of EEG and eye movement (EM) data and the classification of emotions/brain states;
Dual-Modality Watershed Fusion Network for Thyroid Nodule Classification of Dual-View CEUS Video
Li, Rui (Sun Yat-sen University), Lu, Yao (Sun Yat-sen University)
ClassificationConvolutional Neural NetworkOptical FlowImageVideoMultimodalityUltrasound
🎯 What it does: Designed and implemented a Dual-Modal Water Area Fusion Network (DWFN) to classify the benign and malignant thyroid nodules using dual-view CEUS videos and US images.
Dynamic Hybrid Unrolled Multi-Scale Network for Accelerated MRI Reconstruction
Li, Xiao-Xin (Zhejiang University of Technology), Shen, Dinggang (Shanghai United Imaging Intelligence Co., Ltd.)
RestorationRecurrent Neural NetworkTransformerBiomedical DataMagnetic Resonance Imaging
🎯 What it does: This paper proposes a dynamic hybrid stacked network dHUMUS-Net for accelerating magnetic resonance imaging (MRI) reconstruction, which can adaptively select multi-scale Transformer-convolution modules based on the repetitive features of the input image.
Dynamic Position Transformation and Boundary Refinement Network for Left Atrial Segmentation
Xu, Fangqiang (University of Auckland), Zhao, Jichao (University of Auckland)
SegmentationConvolutional Neural NetworkImageBiomedical DataMagnetic Resonance Imaging
🎯 What it does: This paper presents DPBNet, a network for left atrium segmentation that achieves precise segmentation under randomly cropped inputs.
Dynamic Pseudo Label Optimization in Point-Supervised Nuclei Segmentation
Wang, Ziyue (Harbin Institute of Technology), Zhang, Yongbing (Harbin Institute of Technology Shenzhen)
Object DetectionSegmentationConvolutional Neural NetworkContrastive LearningImage
🎯 What it does: Pseudo-labels are generated using point annotations and dynamically optimized, eliminating the need for pixel-level annotations in nuclear segmentation.
Dynamic Single-Pixel Imaging on an Extended Field of View without Warping the Patterns
Maitre, Thomas (Universite Claude Bernard Lyon 1), Sdika, Michaël (Universite Claude Bernard Lyon 1)
RestorationOptical FlowImage
🎯 What it does: This paper addresses the artifact problem caused by motion in dynamic single-pixel imaging (SPI) and proposes a complete decoupling framework that does not convolve distort the light patterns in an extended field of view (FOV);
EchoFM: A View-Independent Echocardiogram Model for the Detection of Pulmonary Hypertension
Fadnavis, Shreyas (Janssen R&D, LLC, a Johnson & Johnson Company), Damasceno, Pablo F. (Janssen R&D, LLC, a Johnson & Johnson Company)
ClassificationRecognitionTransformerContrastive LearningVideoBiomedical DataUltrasound
🎯 What it does: A view-independent Transformer model named EchoFM has been developed for the automatic detection of pulmonary hypertension (PH) from multi-view transthoracic echocardiography (TTE) videos.
EchoMEN: Combating Data Imbalance in Ejection Fraction Regression via Multi-Expert Network
Lai, Song (City University of Hong Kong), Meng, Gaofeng (Chinese Academy of Sciences)
Data-Centric LearningMixture of ExpertsContrastive LearningVideoBiomedical DataUltrasound
🎯 What it does: This paper proposes EchoMEN, a multi-expert network designed to address the data imbalance problem in the regression of ventricular ejection fraction (EF).
EchoNarrator: Generating natural text explanations for ejection fraction predictions
Thomas, Sarina (University of Oslo), Ben-Yosef, Guy (GE Healthcare)
GenerationExplainability and InterpretabilityConvolutional Neural NetworkGraph Neural NetworkLarge Language ModelSupervised Fine-TuningVideoTextUltrasoundChain-of-Thought
🎯 What it does: An end-to-end system was constructed to detect the left ventricle contour using multi-frame GCN from ultrasound videos and directly predict the ejection fraction (EF), followed by generating clinically readable natural language explanations through geometric attributes.
EchoNet-Synthetic: Privacy-preserving Video Generation for Safe Medical Data Sharing
Reynaud, Hadrien (Imperial College London), Kainz, Bernhard (Imperial College London)
GenerationData SynthesisSafty and PrivacyDiffusion modelAuto EncoderVideoBiomedical DataUltrasound
🎯 What it does: A complete end-to-end process has been designed and implemented to generate high-fidelity, long-duration synthetic cardiac ultrasound videos using a latent video diffusion model, with privacy filtering to ensure data de-identification.
EchoTracker: Advancing Myocardial Point Tracking in Echocardiography
Azad, Md Abulkalam (Norwegian University of Science and Technology), Østvik, Andreas (Norwegian University of Science and Technology)
Object TrackingConvolutional Neural NetworkSupervised Fine-TuningImageBiomedical DataUltrasound
🎯 What it does: A learning-based two-stage coarse-fine point tracking model called EchoTracker is proposed, specifically designed for tracking the trajectories of myocardial points in cardiac ultrasound, and utilizes these trajectories to calculate global longitudinal strain (GLS).
Eddeep: Fast eddy-current distortion correction for diffusion MRI with deep learning
Legouhy, Antoine (University College London), Zhang, Hui (AINOSTICS ltd)
Image TranslationRestorationOptimizationComputational EfficiencyConvolutional Neural NetworkSupervised Fine-TuningImageBiomedical DataMagnetic Resonance ImagingDiffusion Tensor Imaging
🎯 What it does: This paper proposes a two-stage process based on deep learning, where a 3D pix2pix translator is first used to convert diffusion-weighted (DW) images with different b-values and gradient directions into images that match the target b-value and have averaged directions, thereby restoring the correspondence between volumes; subsequently, a CNN + spatial transformation network registration model is utilized to estimate and correct gradient current distortions caused by elastic displacement and head motion.
Efficient and Gender-adaptive Graph Vision Mamba for Pediatric Bone Age Assessment
Zhou, Lingyu (Sichuan University), Xu, Xiuyuan (Sichuan University)
ClassificationRecognitionGraph Neural NetworkImage
🎯 What it does: This paper proposes a gender-adaptive graphical visual Mamba network based on raw X-ray images for children's bone age assessment;
Efficient Cortical Surface Parcellation via Full-Band Diffusion Learning at Individual Space
Zhu, Yuanzhuo (Xi'an Jiaotong University), Ma, Jianhua (Pazhou Lab)
SegmentationDiffusion modelBiomedical DataMagnetic Resonance Imaging
🎯 What it does: Cortex-Diffusion is proposed, a lightweight geometric deep network that directly partitions the original cortical surface.
Efficient In-Context Medical Segmentation with Meta-driven Visual Prompt Selection
Wu, Chenwei (University of Michigan), Shen, Liyue (University of Michigan)
SegmentationMeta LearningTransformerReinforcement LearningPrompt EngineeringImageMagnetic Resonance Imaging
🎯 What it does: A meta-learning based visual prompt selection framework MVPS is designed, utilizing a trained Prompt Retriever to achieve efficient context learning for unsupervised medical image segmentation on large visual models.
EgoSurgery-Phase: A Dataset of Surgical Phase Recognition from Egocentric Open Surgery Videos
Fujii, Ryo (Keio University), Kajita, Hiroki (Keio University School of Medicine)
RecognitionTransformerAuto EncoderVideo
🎯 What it does: This paper constructs the first publicly available subjective perspective open surgery video dataset Egosurgery-Phase and proposes the gaze-guided mask autoencoder GGMAE for surgical phase recognition.
EM-Net: Efficient Channel and Frequency Learning with Mamba for 3D Medical Image Segmentation
Chang, Ao (Shenzhen University), Ni, Dong (People's Hospital of Guangxi Zhuang Autonomous Region)
SegmentationConvolutional Neural NetworkImageBiomedical DataComputed Tomography
🎯 What it does: This paper proposes a 3D medical image segmentation framework called EM-Net based on Mamba, which integrates Channel Squeeze-Excitation Blocks (CSRM), Frequency Domain Learning Layers (EFL), and a Mamba decoder to achieve efficient segmentation.
Embracing Massive Medical Data
Chou, Yu-Cheng (Johns Hopkins University), Yuille, Alan (Johns Hopkins University)
SegmentationConvolutional Neural NetworkSupervised Fine-TuningImageBiomedical DataComputed Tomography
🎯 What it does: An online learning framework is proposed, utilizing three mechanisms: linear, dynamic, and selective memory to efficiently process continuous streams of medical data, alleviate catastrophic forgetting, and significantly improve multi-organ and tumor segmentation performance.
Embryo Graphs: Predicting Human Embryo Viability from 3D Morphology
He, Chloe (University College London), Vasconcelos, Francisco (Imperial College London)
ClassificationSegmentationConvolutional Neural NetworkGraph Neural NetworkImageMeshGraphBiomedical Data
🎯 What it does: A non-invasive 3D reconstruction and graphical representation process is proposed, utilizing the generated embryonic graph structure to train a graph neural network to predict the live birth rate of embryos.
EMF-former: An Efficient and Memory-Friendly Transformer for Medical Image Segmentation
Hao, Zhaoquan (East China Normal University), Lu, Yinbin (East China Normal University)
SegmentationTransformerImageBiomedical DataMagnetic Resonance Imaging
🎯 What it does: A lightweight medical image segmentation model EMF-Former based on Transformer is designed and implemented, aiming to significantly reduce the number of parameters, computational complexity, and memory usage while ensuring segmentation accuracy.
Enable the Right to be Forgotten with Federated Client Unlearning in Medical Imaging
Deng, Zhipeng (Hong Kong University of Science and Technology), Chen, Hao (Harvard Medical School)
Federated LearningSafty and PrivacyContrastive LearningImageBiomedical Data
🎯 What it does: Proposed the Federated Client Unlearning (FCU) framework, which implements the 'right to be forgotten' for clients in medical image federated learning;
Enabling Text-free Inference in Language-guided Segmentation of Chest X-rays via Self-guidance
Ye, Shuchang (University of Sydney), Kim, Jinman (University of Sydney)
Object DetectionSegmentationConvolutional Neural NetworkLarge Language ModelImageMultimodality
🎯 What it does: A self-guided segmentation framework SGSeg is proposed, which uses language guidance only during the training phase, allowing for infection area segmentation of chest X-ray images without text during inference.
Endo-4DGS: Endoscopic Monocular Scene Reconstruction with 4D Gaussian Splatting
Huang, Yiming (Chinese University of Hong Kong), Ren, Hongliang (Chinese University of Hong Kong)
Depth EstimationGaussian SplattingImage
🎯 What it does: This paper proposes Endo-4DGS, which utilizes 4D Gaussian splatting to achieve endoscopic real-time variable scene reconstruction.
EndoDAC: Efficient Adapting Foundation Model for Self-Supervised Depth Estimation from Any Endoscopic Camera
Cui, Beilei (Chinese University of Hong Kong), Ren, Hongliang (Chinese University of Hong Kong)
Depth EstimationTransformerOptical FlowVideo
🎯 What it does: Proposed the EndoDAC framework, which efficiently adapts deep foundational models based on Vision Transformer for monocular depth estimation of any endoscopic camera in a self-supervised manner.
EndoFinder: Online Image Retrieval for Explainable Colorectal Polyp Diagnosis
Yang, Ruijie (Fudan University), Wang, Shuo (Fudan University)
RetrievalExplainability and InterpretabilityTransformerContrastive LearningImage
🎯 What it does: The EndoFinder framework is proposed, utilizing content retrieval to achieve interpretable colon polyp diagnosis.
EndoGSLAM: Real-Time Dense Reconstruction and Tracking in Endoscopic Surgeries using Gaussian Splatting
Wang, Kailing (Shanghai Jiao Tong University), Shen, Wei (Zhejiang University)
Object TrackingDepth EstimationOptimizationGaussian SplattingSimultaneous Localization and MappingVideo
🎯 What it does: Proposes EndoGSLAM, a real-time dense SLAM framework for endoscopic surgery, achieving online camera tracking, high-quality tissue reconstruction, and real-time visualization.
Endora: Video Generation Models as Endoscopy Simulators
Li, Chenxin (Chinese University of Hong Kong), Yuan, Yixuan (Southern University of Science and Technology)
GenerationData SynthesisTransformerDiffusion modelAuto EncoderVideoBiomedical DataBenchmark
🎯 What it does: Endora, a space-time Transformer-based endoscopic video generator, has been developed and made publicly available. It can generate realistic and coherent endoscopic simulation videos for medical education, robotic surgery training, and data augmentation.
EndoSelf: Self-Supervised Monocular 3D Scene Reconstruction of Deformable Tissues with Neural Radiance Fields on Endoscopic Videos
Li, Wenda (Nagoya University), Mori, Kensaku (Aichi Institute of Technology)
RestorationGenerationNeural Radiance FieldImageVideo
🎯 What it does: A self-supervised monocular 3D scene reconstruction framework based on NeRF, called EndoSelf, is proposed, which completes the 3D reconstruction of deformable tissues using endoscopic video without depth supervision.
EndoSparse: Real-Time Sparse View Synthesis of Endoscopic Scenes using Gaussian Splatting
Li, Chenxin (Chinese University of Hong Kong), Yuan, Yixuan (Southern University of Science and Technology)
RestorationData SynthesisDiffusion modelGaussian SplattingImage
🎯 What it does: The research focuses on real-time sparse view synthesis and reconstruction of endoscopic scenes using 3D Gaussian light scattering from a sparse perspective.
EndoUIC: Promptable Diffusion Transformer for Unified Illumination Correction in Capsule Endoscopy
Bai, Long (Chinese University of Hong Kong), Ren, Hongliang (Chinese University of Hong Kong)
RestorationSegmentationTransformerPrompt EngineeringDiffusion modelImage
🎯 What it does: A promptable diffusion transformer EndoUIC is proposed for unified illumination correction of wireless capsule endoscopy images.
Energy-Based Controllable Radiology Report Generation with Medical Knowledge
Hou, Zeyi (Beijing University of Posts and Telecommunications), Zhou, Xiuzhuang (University of Science and Technology of China)
GenerationTransformerReinforcement LearningPrompt EngineeringTextBiomedical DataMagnetic Resonance Imaging
🎯 What it does: This paper proposes an energy model-controlled radiology report generation method (ECRG), which incorporates pre-trained expert systems or medical knowledge during the inference phase to impose energy constraints on the report generation process, enabling controllable generation.
Enhanced Scale-aware Depth Estimation for Monocular Endoscopic Scenes with Geometric Modeling
Wei, Ruofeng (Chinese University of Hong Kong), Dou, Qi (Huazhong University of Science and Technology)
Depth EstimationConvolutional Neural NetworkVideo
🎯 What it does: An end-to-end image-based method is proposed, utilizing the geometric model of surgical instruments to achieve scale-aware depth estimation from monocular endoscopic images.
Enhanced-quickDWI: Achieving equivalent clinical quality by denoising heavily sub-sampled diffusion-weighted imaging data
Zormpas-Petridis, Konstantinos (Fondazione Policlinico Universitario Agostino Gemelli IRCCS), Blackledge, Matthew D. (Institute of Cancer Research)
RestorationConvolutional Neural NetworkImageBiomedical DataMagnetic Resonance Imaging
🎯 What it does: A deep learning model called Enhanced-quickDWI was developed for denoising single average DWI data of the whole body and abdominal pelvic region, allowing a 5-minute scan to achieve quality equivalent to that of clinical multi-average images.
Enhancing Federated Learning Performance Fairness via Collaboration Graph-based Reinforcement Learning
Xia, Yuexuan (Northwestern Polytechnical University), Xia, Yong (Northwestern Polytechnical University)
Federated LearningGraph Neural NetworkReinforcement LearningBiomedical DataMagnetic Resonance Imaging
🎯 What it does: A collaborative graph-based reinforcement learning aggregation strategy, FedGraphRL, is proposed to address the performance fairness issue in federated learning.
Enhancing Gait Video Analysis in Neurodegenerative Diseases by Knowledge Augmentation in Vision Language Model
Wang, Diwei (University of Strasbourg), Seo, Hyewon (University of Strasbourg)
ClassificationRecognitionPose EstimationTransformerPrompt EngineeringVision Language ModelContrastive LearningVideoMultimodalityAlzheimer's Disease
🎯 What it does: Developed a knowledge-enhanced visual language model to assess the diagnostic group of neurodegenerative diseases and gait disorders using monocular gait videos.
Enhancing Gene Expression Prediction from Histology Images with Spatial Transcriptomics Completion
Mejia, Gabriel (Universidad de los Andes), Arbeláez, Pablo (Universidad de los Andes)
Graph Neural NetworkTransformerImageBiomedical Data
🎯 What it does: This paper constructs 26 standardized Visium Spatial Transcriptomics datasets (SpaRED) and proposes a Transformer-based reference-independent missing value imputation model, SpaCKLE, to enhance gene expression prediction performance.
Enhancing Human-Computer Interaction in Chest X-ray Analysis using Vision and Language Model with Eye Gaze Patterns
Kim, Yunsoo (University College London), Wu, Honghan (University College London)
TransformerVision Language ModelImageMultimodalityBiomedical DataComputed Tomography
🎯 What it does: This paper inputs the eye-tracking heatmaps of radiologists along with chest X-ray images into a visual-language model to enhance the accuracy of multi-task diagnosis.
Enhancing Label-efficient Medical Image Segmentation with Text-guided Diffusion Models
Feng, Chun-Mei (Agency for Science, Technology and Research)
SegmentationConvolutional Neural NetworkDiffusion modelImageTextMultimodalityBiomedical DataMagnetic Resonance Imaging
🎯 What it does: A medical image segmentation framework called TextDiff based on diffusion models is proposed, which utilizes diagnostic text annotations to enhance visual semantic representation, thereby achieving efficient segmentation with few labeled samples.
Enhancing Model Generalisability through Sampling Diverse and Balanced Retinal Images
Zhou, Tianfeng (Central South University), Zhou, Yukun (University College London)
ClassificationSegmentationConvolutional Neural NetworkTransformerImage
🎯 What it does: A two-stage sampling pipeline based on foundational model features (DataDIVA) is proposed, which enhances the model's generalization performance on both internal and external datasets through diversity and balanced sampling on retinal images.
Enhancing New Multiple Sclerosis Lesion Segmentation via Self-supervised Pre-training and Synthetic Lesion Integration
Tahghighi, Peyman (University of Calgary), Komeili, Amin (University of Calgary)
SegmentationData SynthesisConvolutional Neural NetworkSupervised Fine-TuningImageBiomedical DataMagnetic Resonance Imaging
🎯 What it does: By using self-supervised mask pre-training and synthetic lesion augmentation, the accuracy of new lesion segmentation in multiple sclerosis is improved; training is conducted on a VNet-based segmentation model using both real and synthetic lesions as input; model parameters are pre-trained through self-supervised tasks, and consistency loss and boundary loss are added in the segmentation task; multi-fold cross-validation evaluation is implemented.
Enhancing Spatiotemporal Disease Progression Models via Latent Diffusion and Prior Knowledge
Puglisi, Lemuel (University of Catania), Ravì, Daniele (University College London)
SegmentationGenerationData SynthesisDiffusion modelAuto EncoderImageBiomedical DataMagnetic Resonance ImagingAlzheimer's Disease
🎯 What it does: Proposed and implemented BrLP, a 3D brain MRI spatiotemporal disease progression prediction framework utilizing latent diffusion models, ControlNet, auxiliary models, and LAS technology.
Enhancing Whole Slide Image Classification with Discriminative and Contrastive Learning
Liang, Peixian (University of Pennsylvania), Fan, Yong (Indiana University)
ClassificationTransformerContrastive LearningImage
🎯 What it does: In digital pathology, for the classification task of whole slide images (WSI), this paper proposes an end-to-end dual learning framework—DC-WSI, which improves classification performance by first selecting representative patches and then combining discriminative learning with contrastive learning.
Ensemble of Prior-guided Expert Graph Models for Survival Prediction in Digital Pathology
Ramanathan, Vishwesh (University of Toronto), Martel, Anne L. (University of Toronto)
ClassificationSegmentationGraph Neural NetworkImageBiomedical Data
🎯 What it does: A directed edge attribute tissue subgraph is constructed using tissue and TIL segmentation priors, and a specialized graph attention network is trained on each subgraph. Finally, the expert models are linearly integrated to obtain the final survival prediction.
Ensembled Cold-Diffusion Restorations for Unsupervised Anomaly Detection
Naval Marimont, Sergio (City University of London), Tarroni, Giacomo (CitAI Research Centre)
RestorationSegmentationAnomaly DetectionDiffusion modelImageBiomedical DataMagnetic Resonance Imaging
🎯 What it does: This paper presents DISYRE v2, an unsupervised anomaly detection framework that combines cold diffusion recovery and synthetic anomaly generation for abnormal localization in brain MRI.
Epicardium Prompt-guided Real-time Cardiac Ultrasound Frame-to-volume Registration
Lei, Long (Chinese University of Hong Kong), Heng, Pheng-Ann (Chinese University of Hong Kong)
SegmentationPose EstimationConvolutional Neural NetworkPrompt EngineeringImageBiomedical DataUltrasound
🎯 What it does: A lightweight end-to-end cardiac ultrasound image frame-volume registration network CU-Reg has been designed and implemented to register real-time two-dimensional ultrasound images with preoperative three-dimensional ultrasound volumes, providing a complete surgical navigation view.
Epileptic Seizure Detection in SEEG Signals using a Unified Multi-scale Temporal-Spatial-Spectral Transformer Model
Li, Zhuoyi (Northwestern Polytechnical University), Zhang, Tuo (Northwestern Polytechnic University)
ClassificationAnomaly DetectionTransformerTime SeriesBiomedical Data
🎯 What it does: A unified multi-scale spatio-temporal frequency domain Transformer framework (CE-TSS-Transformer) is proposed for detecting long-term SEEG signals of epilepsy seizures, and a novel LTSZ epilepsy seizure SEEG dataset is constructed.
ESPA: An Unsupervised Harmonization Framework via Enhanced Structure Preserving Augmentation
Eshaghzadeh Torbati, Mahbaneh (University of Pittsburgh), Hwang, Seong Jae (Mediwhale)
Image HarmonizationRestorationGenerative Adversarial NetworkImageMagnetic Resonance Imaging
🎯 What it does: An unsupervised image harmonization framework called ESPA is proposed, which generates matched data by simulating scanner effects and using structure-preserving enhancement methods to achieve debiasing of multi-scanner images.
Estimating Neural Orientation Distribution Fields on High Resolution Diffusion MRI Scans
Dwedari, Mohammed Munzer (Harvard Medical School), Rathi, Yogesh (Technical University of Munich)
Diffusion modelBiomedical DataMagnetic Resonance ImagingFibre Orientation DistributionDiffusion Tensor Imaging
🎯 What it does: This paper proposes an implicit neural representation method based on grid hash encoding (HashEnc) for continuously estimating the orientation distribution function (ODF) field in high-resolution diffusion magnetic resonance scans.
Estimation and Analysis of Slice Propagation Uncertainty in 3D Anatomy Segmentation
Nihalaani, Rachaell (University of Utah), Elhabian, Shireen Y. (University of Utah)
SegmentationBiomedical DataMagnetic Resonance ImagingComputed Tomography
🎯 What it does: In the task of 3D medical image segmentation with no annotations or only single-slice annotations, five methods of empirical uncertainty quantification (UQ) based on Bayesian or ensemble approaches were introduced and evaluated to enhance the credibility and accuracy of self-supervised slice propagation segmentation models.
Evaluating the Fairness of Neural Collapse in Medical Image Classification
Mouheb, Kaouther (Erasmus University Medical Center), Bron, Esther E. (Hong Kong University of Science and Technology)
ClassificationImageBiomedical Data
🎯 What it does: The study investigates how deep learning models gradually converge to the Neural Collapse (NC) state during training in the presence of label bias (such as misdiagnosis of a certain group), and the impact of NC on the fairness of medical image classification.
Evaluating the Quality of Brain MRI Generators
Wu, Jiaqi (Stanford University), Pohl, Kilian M. (Stanford)
SegmentationGenerationDiffusion modelGenerative Adversarial NetworkImageBiomedical DataMagnetic Resonance ImagingAlzheimer's Disease
🎯 What it does: Evaluate the quality of brain MRI generators and propose a unified framework for preprocessing, implementation, and anatomical plausibility assessment.
Evidential Concept Embedding Models: Towards Reliable Concept Explanations for Skin Disease Diagnosis
Gao, Yibo (Fudan University), Zhuang, Xiahai (Fudan University)
ClassificationExplainability and InterpretabilityVision Language ModelImageBiomedical Data
🎯 What it does: A concept bottleneck model based on evidence reasoning (evi-CEM) is proposed for skin disease diagnosis; simultaneously, a concept correction mechanism (using CAV to correct concept biases generated by VLM) and an uncertainty-based intervention strategy are designed.