Is this you? Claim this profile to correct it, add a bio and choose the work people see first.
Claim this profileWorks346 from public data
- Diagnostic Assessment of Deep Learning Algorithms for Detection of Lymph Node Metastases in Women With Breast Cancer3,315
In the setting of a challenge competition, some deep learning algorithms achieved better diagnostic performance than a panel of 11 pathologists participating in a simulation exercise designed to mimic routine pathology workflow; algorithm performance was comparable with an expert pathologist interpreting whole-slide images without time constraints.
- 2,507
- 1,593
- Validation, comparison, and combination of algorithms for automatic detection of pulmonary nodules in computed tomography images: The LUNA16 challenge1,414
The LUNA16 challenge is described, an objective evaluation framework for automatic nodule detection algorithms using the largest publicly available reference database of chest CT scans, the LIDC‐IDRI data set, and the results so far are presented.
- 1,146
- VoxResNet: Deep voxelwise residual networks for brain segmentation from 3D MR images790
An auto‐context version of the VoxResNet is proposed by combining the low‐level image appearance features, implicit shape information, and high‐level context together for further improving the segmentation performance, and achieved the best performance in the 2013 MICCAI MRBrainS challenge.
- 692
- FedDG: Federated Domain Generalization on Medical Image Segmentation via Episodic Learning in Continuous Frequency Space650
This paper points out and solves a novel problem setting of federated domain generalization (FedDG), which aims to learn a federated model from multiple distributed source domains such that it can directly generalize to unseen target domains.
- 3D deeply supervised network for automated segmentation of volumetric medical images627
The proposed 3D DSN is capable of conducting volume‐to‐volume learning and inference, which can eliminate redundant computations and alleviate the risk of over‐fitting on limited training data, and the3D deep supervision mechanism can effectively cope with the optimization problem of gradients vanishing or exploding when training a 3D deep model.
- 595
- DCAN: Deep contour-aware networks for object instance segmentation from histology images518
A novel deep contour‐aware network (DCAN) under a unified multi‐task learning framework for more accurate detection and segmentation of objects of interest from histology images is proposed.
- Unsupervised Bidirectional Cross-Modality Adaptation via Deeply Synergistic Image and Feature Alignment for Medical Image Segmentation450
This work presents a novel unsupervised domain adaptation framework, named as Synergistic Image and Feature Alignment (SIFA), to effectively adapt a segmentation network to an unlabeled target domain and outperforms the state-of-the-art domain adaptation approaches by a large margin.
- 423
- 412
- Federated learning enables big data for rare cancer boundary detection398
The largest FL study to-date is presented, involving data from 71 sites across 6 continents, to generate an automatic tumor boundary detector for the rare disease of glioblastoma, reporting the largest such dataset in the literature.
- Synergistic Image and Feature Adaptation: Towards Cross-Modality Domain Adaptation for Medical Image Segmentation388
This paper presents a novel unsupervised domain adaptation framework, called Synergistic Image and Feature Adaptation (SIFA), to effectively tackle the problem of domain shift, and extensively validated the method with a challenging application of cross-modality medical image segmentation of cardiac structures.
- 3D Deeply Supervised Network for Automatic Liver Segmentation from CT Volumes375
A novel 3D deeply supervised network (3D DSN) is presented which takes advantage of a fully convolutional architecture which performs efficient end-to-end learning and inference and introduces a deep supervision mechanism during the learning process to combat potential optimization difficulties.
- MILD-Net: Minimal information loss dilated network for gland instance segmentation in colon histology images372
A fully convolutional neural network is proposed that counters the loss of information caused by max‐pooling by re‐introducing the original image at multiple points within the network, and introduces random transformations during test time for an enhanced segmentation result that simultaneously generates an uncertainty map, highlighting areas of ambiguity.
- Harmonizing Transferability and Discriminability for Adapting Object Detectors348
A Hierarchical Transferability Calibration Network (HTCN) that hierarchically (local-region/image/instance) calibrates the transferability of feature representations for harmonizing transferability and discriminability is proposed.
- Deep Learning for Automated Contouring of Primary Tumor Volumes by MRI for Nasopharyngeal Carcinoma345
The AI contouring tool improved primary gross tumor contouring accuracy of nasopharyngeal carcinoma, which could have a positive impact on tumor control and patient survival.
- 340
- Unsupervised Cross-Modality Domain Adaptation of ConvNets for Biomedical Image Segmentations with Adversarial Loss336
An unsupervised domain adaptation framework with adversarial learning for cross-modality biomedical image segmentations for cardiac structures segmentations is proposed, and very promising results are achieved.
- 326
- Shape-Aware Meta-learning for Generalizing Prostate MRI Segmentation to Unseen Domains295
A novel shape-aware meta-learning scheme is presented to improve the model generalization in prostate MRI segmentation, and outperforms many state-of-the-art generalization methods consistently across all six settings of unseen domains.
- Federated deep learning for detecting COVID-19 lung abnormalities in CT: a privacy-preserving multinational validation study282
The federated learning algorithms are explored to develop a privacy-preserving AI model for COVID-19 medical image diagnosis with good generalization capability on unseen multinational datasets.
- Mitosis Detection in Breast Cancer Histology Images via Deep Cascaded Networks270
A fast and accurate method to detect mitosis is proposed by designing a novel deep cascaded convolutional neural network, which is composed of two components: a coarse retrieval model to identify and locate the candidates of mitosis while preserving a high sensitivity and a fine discrimination model utilizing knowledge transferred from cross-domain.
- Automatic 3D Cardiovascular MR Segmentation with Densely-Connected Volumetric ConvNets249
A novel densely-connected volumetric convolutional neural network to automatically segment the cardiac and vascular structures from 3D cardiac MR images, referred as DenseVoxNet, and it is shown that this network can achieve better performance than other 3D ConvNets but with fewer parameters.
- Autonomous environment-adaptive microrobot swarm navigation enabled by deep learning-based real-time distribution planning248
This work could introduce computational intelligence to micro-/nanorobot swarms, enabling them to autonomously make appropriate decisions during navigation in unstructured environments, and presents a real-time autonomous distribution planning method based on deep learning that controls both the shape and position of the swarm.
- 247
- CIA-Net: Robust Nuclei Instance Segmentation with Contour-Aware Information Aggregation243
A novel deep neural network, namely Contour-aware Informative Aggregation Network (CIA-Net) with multi-level information aggregation module between two task-specific decoders that leverages the merit of spatial and texture dependencies between nuclei and contour by bi-directionally aggregating task- specific features.
- Multi-task recurrent convolutional network with correlation loss for surgical video analysis231
A novel method by developing a multi-task recurrent convolutional network with correlation loss (MTRCNet-CL) to exploit their relatedness to simultaneously boost the performance of both surgical tool presence detection and phase recognition.
- Robust Multimodal Brain Tumor Segmentation via Feature Disentanglement and Gated Fusion229
This work proposes a novel multimodal segmentation framework which is robust to the absence of imaging modalities, and achieves outstanding robustness under various missing modality(ies) situations, significantly exceeding the state-of-the-art method on whole tumor segmentation.
- Unpaired Multi-Modal Segmentation via Knowledge Distillation222
This work proposes a novel learning scheme for unpaired cross-modality image segmentation, with a highly compact architecture achieving superior segmentation accuracy, and introduces a novel loss term inspired by knowledge distillation.
- HarmoFL: Harmonizing Local and Global Drifts in Federated Learning on Heterogeneous Medical Images220
This work proposes to mitigate the local update drift by normalizing amplitudes of images transformed into the frequency domain to mimic a unified imaging setting, and introduces a new harmonizing framework called HarmoFL, which outperforms a set of recent state of theart methods with promising convergence behavior.
- Contrastive Cross-Site Learning With Redesigned Net for COVID-19 CT Classification217
A novel joint learning framework to perform accurate COVID-19 identification by effectively learning with heterogeneous datasets with distribution discrepancy is proposed and a powerful backbone is built by redesigning the recently proposed CO VID-Net in aspects of network architecture and learning strategy to improve the prediction accuracy and learning efficiency.
- PnP-AdaNet: Plug-and-Play Adversarial Domain Adaptation Network at Unpaired Cross-Modality Cardiac Segmentation217
A plug-and-play adversarial domain adaptation network (PnP-AdaNet) for adapting segmentation networks between different modalities of medical images, e.g., MRI and CT, which outperforms many state-of-the-art unsupervised domain adaptation approaches on the same dataset.
- 215
- Trans-SVNet: Accurate Phase Recognition from Surgical Videos via Hybrid Embedding Aggregation Transformer183
This paper introduces, for the first time in surgical workflow analysis, Transformer to reconsider the ignored complementary effects of spatial and temporal features for accurate surgical phase recognition in hybrid embedding aggregation Transformer.
- Automatic Fetal Ultrasound Standard Plane Detection Using Knowledge Transferred Recurrent Neural Networks177
This work proposes a joint learning framework with knowledge transfer across multi-tasks to address the insufficiency issue of limited training data and explores spatio-temporal feature learning with a novel knowledge transferred recurrent neural network T-RNN.
- Automated Pulmonary Nodule Detection via 3D ConvNets with Online Sample Filtering and Hybrid-Loss Residual Learning173
A novel framework with 3D convolutional networks (ConvNets) for automated detection of pulmonary nodules from low-dose CT scans, which is a challenging yet crucial task for lung cancer early diagnosis and treatment is proposed.
- Semantic-Aware Generative Adversarial Nets for Unsupervised Domain Adaptation in Chest X-Ray Segmentation172
A novel unsupervised domain adaptation approach for segmentation tasks by designing semantic-aware generative adversarial networks (GANs) that transform the test image into the appearance of source domain, with the semantic structural information being well preserved.
- 171
- 166
- Source-Free Domain Adaptive Fundus Image Segmentation with Denoised Pseudo-Labeling163
This paper presents a novel denoised pseudo-labeling method, which effectively makes use of the source model and unlabeled target data to promote model self-adaptation from pseudo labels, and introduces two complementary pixel-level and class-level denoising schemes to reduce noisy pseudo labels and select reliable ones to enhance the pseudo- labeling efficacy.
- Weakly supervised 3D deep learning for breast cancer classification and localization of the lesions in MR images158
The usefulness of 3D deep learning‐based classification of breast cancer and malignancy localization from MRI has been reported and can potentially be very useful in the clinical domain and aid radiologists in breast cancer diagnosis.
- 3D multi-scale FCN with random modality voxel dropout learning for Intervertebral Disc Localization and Segmentation from Multi-modality MR Images158
A novel multi‐scale and modality dropout learning framework to locate and segment IVDs from four‐modality MR images and presents a random modality voxel dropout strategy which alleviates the co‐adaption issue and increases the discriminative capability of the network.
- 3DSAM-adapter: Holistic adaptation of SAM from 2D to 3D for promptable tumor segmentation155
This paper proposes a novel adaptation method for transferring SAM from 2D to 3D for promptable medical image segmentation through a holistically designed scheme for architecture modification, which can outperform domain state-of-the-art medical image segmentation models and interactive segmentation models.
- Temporal Memory Relation Network for Workflow Recognition From Surgical Video153
A novel end-to-end temporal memory relation network (TMRNet) for relating long-range and multi-scale temporal patterns to augment the present features, and introduces a non-local bank operator to attentively relate the past to the present.
- Incorporating Temporal Prior from Motion Flow for Instrument Segmentation in Minimally Invasive Surgery Video148
This paper proposes a novel framework to leverage instrument motion information, by incorporating a derived temporal prior to an attention pyramid network for accurate segmentation, and demonstrates a promising potential for reducing annotation cost in the clinical practice.
- Comparative validation of machine learning algorithms for surgical workflow and skill analysis with the HeiChole benchmark142
Surgical workflow and skill analysis are promising technologies to support the surgical team, but there is still room for improvement, as shown by the comparison of machine learning algorithms.
- Category-Level 6D Object Pose Estimation via Cascaded Relation and Recurrent Reconstruction Networks139
A novel cascaded relation network is dedicated for advanced representation learning to explore the complex and informative relations among instance RGB image, instance point cloud and category shape prior and a recurrent reconstruction network for iterative residual refinement is designed to progressively improve the reconstruction and correspondence estimations from coarse to fine.
- 133
- Robust Learning at Noisy Labeled Medical Images: Applied to Skin Lesion Classification129
This paper proposes an effective iterative learning framework for noisy-labeled medical image classification, to combat the lacking of high quality annotated medical data, and designs a sample re-weighting strategy to preserve the usefulness of correctly- labeled hard samples.
- Transformer-empowered Multi-scale Contextual Matching and Aggregation for Multi-contrast MRI Super-resolution123
A novel network of innovative Transformer-empowered multi-scale contextual matching and aggregation techniques is proposed, which outperforms state-of-the-art approaches and has great potential to be applied in clinical practice.
- 122
- SurRoL: An Open-source Reinforcement Learning Centered and dVRK Compatible Platform for Surgical Robot Learning119
The designed SurRoL integrates a user-friendly RL library for algorithm development and a real-time physics engine, which is able to support more PSM/ECM scenarios and more realistic physical interactions, and achieves better transferability in the real world.
- Improving RetinaNet for CT Lesion Detection with Dense Masks from Weak RECIST Labels119
This work proposes a highly accurate and efficient one-stage lesion detector, by re-designing a RetinaNet to meet the particular challenges in medical imaging, and optimize the anchor configurations using a differential evolution search algorithm.
- 114
- 113
- Fair Federated Medical Image Segmentation via Client Contribution Estimation109
This work proposes a novel method to optimize both types of fairness simultaneously, federated training via contribution estimation (FedCE), i.e., using estimation as global model aggregation weights.
- Federated Semi-supervised Medical Image Classification via Inter-client Relation Matching109
This paper studies a practical yet challenging FL problem, named FSSL, which aims to learn a federated model by jointly utilizing the data from both labeled and unlabeled clients, and presents a novel approach for this problem, which improves over traditional consistency regularization mechanism with a new inter-client relation matching scheme.
- Multi-Site Infant Brain Segmentation Algorithms: The iSeg-2019 Challenge106
8 top-ranked methods in the iSeg-2019 challenge were reviewed by detailing their pipelines/implementations, presenting experimental results, and evaluating performance across different sites in terms of whole brain, regions of interest, and gyral landmark curves, finding that multi-site consistency is still an open issue.
- H-DenseUNet: Hybrid Densely Connected UNet for Liver and Liver Tumor Segmentation from CT Volumes106
A novel hybrid densely connected UNet (H-Dense UNet), which consists of a 2D DenseUNet for efficiently extracting intra-slice features and a 3D counterpart for hierarchically aggregating volumetric contexts under the spirit of the auto-context algorithm for liver and tumor segmentation.
- An Encoder-Decoder Neural Network With 3D Squeeze-and-Excitation and Deep Supervision for Brain Tumor Segmentation89
This work proposed an encoder-decoder neural network, i.e. deep supervised 3D Squeeze-and-Excitation V-Net (DSSE-V-Net) to segment brain tumors automatically, which outperformed both 3D U-Net and modified V- net, and obtained highly competitive performance compared with those methods winning in the BraTS 2017 challenge.
- Triplet attention and dual-pool contrastive learning for clinic-driven multi-label medical image classification87
A hybrid label is designed to improve the flexibility of MLC methods and alleviate the sample imbalance problem, and the hybrid label can also be used to put unseen diseases in practical use.
- E-DSSR: Efficient Dynamic Surgical Scene Reconstruction with Transformer-Based Stereoscopic Depth Perception86
An efficient reconstruction pipeline for highly dynamic surgical scenes that runs at 28 fps is presented and can recover the scene obstructed by the surgical tool and handle the movement of camera in realistic surgical scenarios effectively at real-time speed.
- Evaluation and comparison of 3D intervertebral disc localization and segmentation methods for 3D T2 MR data: A grand challenge85
Results achieved by the best algorithms in this study set new frontiers for IVD localization and segmentation from MR data.
- 3-D RoI-Aware U-Net for Accurate and Efficient Colorectal Tumor Segmentation84
A novel encoder–decoder-based framework for 3-D whole volume segmentation and a dice-based multitask hybrid loss function to facilitate the global-to-local learning procedure and enhance contour detail richness are proposed.
- ScanNet: A Fast and Dense Scanning Framework for Metastastic Breast Cancer Detection from Whole-Slide Image84
A novel framework by leveraging fully convolutional networks for efficient inference to meet the speed requirement for clinical practice, while reconstructing dense predictions under different offsets for ensuring accurate detection on both microand macro-metastases is proposed.
- Federated Domain Generalization for Image Recognition via Cross-Client Style Transfer81
This paper proposes a novel domain generalization method for image recognition under federated learning through cross-client style transfer (CCST) without exchanging data samples, which can lead to more uniform distributions of source clients, and make each local model learn to fit the image styles of all the clients to avoid the different model biases.
- IOP-FL: Inside-Outside Personalization for Federated Medical Image Segmentation77
This work designs a new test-time routing scheme using the consistency loss with a shape constraint to dynamically incorporate the models, given the distribution information conveyed by the test data, demonstrating the potential of the IOP-FL scheme for clinical practice.
- 77
- Surgical embodied intelligence for generalized task autonomy in laparoscopic robot-assisted surgery74
This study further addresses surgical task automation, enabling zero-shot transfer of simulation-trained policies to real-world scenarios, and introduced an open-source surgical embodied intelligence simulator for an interactive environment to develop reinforcement learning methods for minimally invasive surgical robots.
- Human-in-the-Loop Embodied Intelligence With Interactive Simulation Environment for Surgical Robot Learning74
This work establishes a platform based on the previously released SurRoL simulator with several new features co-developed to allow high-quality human interaction via an input device for surgical robot learning, and proves effectiveness of incorporating human factors in embodied intelligence through the use of human demonstrations and reinforcement learning.
- 74
- 71
- Webthetics: Quantifying webpage aesthetics with deep learning69
A novel method using deep learning to automatically compute and quantify webpage aesthetics, which significantly outperforms previous method using hand-crafted features such as colorfulness and complexity and indicates that it can serve as an effective and efficient means for providing objective aesthetics evaluation during the design process.
- 67
- 67
- 3D Fully Convolutional Networks for Intervertebral Disc Localization and Segmentation63
2D fully convolutional networks are extended into a 3D variant with end-to-end learning and inference, where voxel-wise predictions are generated and it is corroborated that 3D FCN can achieve a higher localization and segmentation accuracy than 2DFCN, which demonstrates the significance of volumetric information when confronting 3D localization and segmentsation tasks.
- DLTTA: Dynamic Learning Rate for Test-Time Adaptation on Cross-Domain Medical Images62
A novel dynamic learning rate adjustment method for test-time adaptation, called DLTTA, which dynamically modulates the amount of weights update for each test image to account for the differences in their distribution shift.
- Region-focused multi-view transformer-based generative adversarial network for cardiac cine MRI reconstruction60
A region-focused multi-view transformer-based generative adversarial network for cardiac cine MRI reconstruction that outperforms state-of-the-art methods and introduces a cardiac region detection loss for improving the reconstruction quality of the cardiac region.
- Automatic Gesture Recognition in Robot-assisted Surgery with Reinforcement Learning and Tree Search59
A framework based on reinforcement learning and tree search for joint surgical gesture segmentation and classification that consistently outperforms the existing methods on the suturing task of JIGSAWS dataset in terms of accuracy, edit score and F1 score.
- Video Dehazing via a Multi-Range Temporal Alignment Network with Physical Prior58
A memory-based physical prior guidance module to encode the prior-related features into long-range memory and a multi-range scene radiance recovery module to capture space-time dependencies in multiple space- time ranges, which helps to effectively aggregate temporal information from adjacent frames.
- MIcro-surgical anastomose workflow recognition challenge report56
For high levels of granularity, the best models had a recognition rate that may be sufficient for applications such as prediction of remaining surgical time, however, for activities, the recognition rate was still low for applications that can be employed clinically.
- Learning Motion Flows for Semi-supervised Instrument Segmentation from Robotic Surgical Video55
This paper proposes a dual motion based method to wisely learn motion flows for segmentation enhancement by leveraging temporal dynamics, and designs a flow predictor to derive the motion for jointly propagating the frame-label pairs given the current labeled frame.
- SFCN-OPI: Detection and Fine-Grained Classification of Nuclei Using Sibling FCN With Objectness Prior Interaction55
A novel method of sibling fully convolutional network with prior objectness interaction (called SFCN-OPI) to tackle the two tasks simultaneously and interactively using a unified end-to-end framework and has outperformed the state-of-the-art methods by a large margin.
- Self-Ensembling Co-Training Framework for Semi-Supervised COVID-19 CT Segmentation54
A self-ensembled co-training framework, which is trained by limited labeled data and large-scale unlabeled data, to automatically extract COVID lesions from CT scans is proposed and achieves better performance in the case of only 4 labeled CT scans compared to the state-of-the-art semi-supervised segmentation networks.
- IRNet: Instance Relation Network for Overlapping Cervical Cell Segmentation54
A novel Instance Relation Network (IRNet) for robust overlapping cell segmentation by exploring instance relation interaction is proposed to construct the cell association matrix for transferring information among individual cell-instance features.
- 53
- 53
- Stereo Dense Scene Reconstruction and Accurate Localization for Learning-Based Navigation of Laparoscope in Minimally Invasive Surgery52
A learning-driven framework is proposed, in which an image-guided laparoscopic localization with 3D reconstructions of anatomical structures is obtained and can be potentially extended to the current surgical navigation system.
- Intelligent surgical workflow recognition for endoscopic submucosal dissection with real-time animal study51
The proposed AI-Endo, an intelligent surgical workflow recognition suit, for endoscopic submucosal dissection (ESD) is trained on high-quality ESD cases from an expert endoscopist, covering a decade time expansion and consisting of 201,026 labeled frames.
- Federated Learning in Ocular Imaging: Current Progress and Future Direction51
The concept of FL is introduced, current evidence of FL in ocular imaging is provided, potential challenges as well as future applications are discussed, and potential challenges and future applications of FL are discussed.
- A Dexterous and Compliant (DexCo) Hand Based on Soft Hydraulic Actuation for Human-Inspired Fine In-Hand Manipulation50
Experimental results demonstrate the DexCo hand successfully performs complex fine in-hand manipulation tasks, providing a promising solution for advancing robotic manipulation capabilities toward the human level.
- Deep Fusion Transformer Network with Weighted Vector-Wise Keypoints Voting for Robust 6D Object Pose Estimation49
A novel Deep Fusion Transformer block that can aggregate cross-modality features for improving pose estimation and a novel weighted vector-wise voting algorithm that employs a non-iterative global optimization strategy for precise 3D keypoint localization while achieving near real-time inference.
- DeepKOA: a deep-learning model for predicting progression in knee osteoarthritis using multimodal magnetic resonance images from the osteoarthritis initiative48
The feasibility of DeepKOA was demonstrated in the prediction of KOA progression and the potential responsible structures which may enlighten the future development of more clinically practical methods were identified.
- Relational Graph Learning on Visual and Kinematics Embeddings for Accurate Gesture Recognition in Robotic Surgery48
This work proposes a novel online approach of multi-modal relational graph network (i.e., MRG-Net) to dynamically integrate visual and kinematics information through interactive message propagation in the latent feature space, and demonstrates the effectiveness of the method with state-of-the-art results on the public JIGSAWS dataset.
- An Active Learning Approach for Reducing Annotation Cost in Skin Lesion Analysis48
This paper presents a novel active learning framework for cost-effective skin lesion analysis, to effectively select and utilize much fewer labelled samples, while the network can still achieve state-of-the-art performance.
- Demonstration-Guided Reinforcement Learning with Efficient Exploration for Task Automation of Surgical Robot46
Demonstration-guided EXploration (DEX) is introduced, an efficient reinforcement learning algorithm that aims to overcome the exploration problem with expert demonstrations for surgical automation by estimates expert-like behaviors with higher values to facilitate productive interactions and adopts non-parametric regression to enable such guidance at states unobserved in demonstration data.
- Federated Learning from Only Unlabeled Data with Class-Conditional-Sharing Clients45
Federation of unsupervised learning (FedUL), where the unlabeled data are transformed into surrogate labeled data for each of the clients, a modified model is trained by supervised FL, and the wanted model is recovered from the modified model.
- 45
- 43
- Weakly Supervised Learning for Whole Slide Lung Cancer Image Classification43
A weakly supervised approach for fast and effective classification on whole slide lung cancer images that takes advantage of a patch-based fully convolutional network for discriminative block retrieval and context-aware feature selection and aggregation strategies are proposed to generate globally holistic WSI descriptor.
- 43
- 41
- 40
- 39
- 39
- 39
- RecolorNeRF: Layer Decomposed Radiance Fields for Efficient Color Editing of 3D Scenes38
RecolorNeRF is presented, a novel user-friendly color editing approach for the neural radiance fields that outperforms baseline methods both quantitatively and qualitatively for color editing even in complex real-world scenes.
- 37
- Trans-SVNet: hybrid embedding aggregation Transformer for surgical workflow analysis37
This paper proposes a hybrid embedding aggregation Transformer, named Trans-SVNet, to effectively interact with the designed spatial and temporal embeddings, by employing spatial embedding to query temporal embedding sequence and jointly optimized by loss objectives from both analysis tasks to leverage their high correlation.
- Cascaded Robust Learning at Imperfect Labels for Chest X-ray Segmentation37
A novel cascaded robust learning framework for chest X-ray segmentation with imperfect annotation at the boundary is presented, which consists of three independent networks, which can effectively learn useful information from peer networks.
- Efficient Deformable Tissue Reconstruction via Orthogonal Neural Plane36
Fast Orthogonal Plane (Forplane), a novel, efficient framework based on neural radiance fields (NeRF) for the reconstruction of deformable tissues, is introduced, which substantially accelerates both the optimization process and the inference process while maintaining or even improving the quality across a variety of non-rigid deformations.
- 36
- Machine Learning and Radiomics Applications in Esophageal Cancers Using Non-Invasive Imaging Methods—A Critical Review of Literature36
This review aims to explore the recent advances and future perspective of the ML technique in the disease management of esophageal cancer patients, with a focus on imaging pertinent to EC patient care.
- Semi-supervised learning with progressive unlabeled data excavation for label-efficient surgical workflow recognition35
A novel two-stage Semi-Supervised Learning method for label-efficient Surgical workflow recognition, named as SurgSSL that surpasses the state-of-the-art semi-supervised methods by a large margin and demonstrates superior reliability for model supervision compared with the label generated by existing methods.
- LRTD: long-range temporal dependency based active learning for surgical workflow recognition35
A novel active learning method for cost-effective surgical video analysis is proposed, which introduces non-local recurrent convolutional network to capture the long-range temporal dependency (LRTD) among continuous frames and forms an intra-clip dependency score to represent the overall dependency within this clip.
- StereoPose: Category-Level 6D Transparent Object Pose Estimation from Stereo Images via Back-View NOCS34
StereoPose is a novel stereo image based framework for category-level object pose estimation, ideally suited for transparent objects, and is equipped with a parallax attention module for stereo feature fusion and an epipolar loss for improving the stereo-view consistency of network predictions.
- Why is the Winner the Best?34
A multicenter study with all 80 benchmarking competitions that were conducted in the scope of IEEE ISBI 2021 and MICCAI 2021 revealed common characteristics of winning solutions: the reflection of the metrics in the method design and the focus on analyzing and handling failure cases.
- DuDoCAF: Dual-Domain Cross-Attention Fusion with Recurrent Transformer for Fast Multi-contrast MR Imaging34
The proposed DuDoCAF scheme with recurrent transformer enables deep andective fusion of features extracted from two modalities, and dual-domain recurrent learning allows the model to restore signals in both k -space and image domains, and hence more comprehensively remove the artifacts.
- Data-driven Holistic Framework for Automated Laparoscope Optimal View Control with Learning-based Depth Perception34
A novel rotation constraint using an affine map to minimize the visual warping problem, and a null-space controller is also embedded into the framework to optimize all types of errors in a unified and decoupled manner.
- Artificial intelligence assisted real-time recognition of intra-abdominal metastasis during laparoscopic gastric cancer surgery33
The artificial intelligence laparoscopic exploration system (AiLES) to recognize IAM lesions with various metastatic extents and locations performed comparably to novice surgeons in IAM recognition and excelled in recognizing tiny and occult lesions.
- 31
- Morphology-aware multi-source fusion–based intracranial aneurysms rupture prediction31
A self-supervised learning method was proposed to mitigate the data-hungry issue of deep learning, enabling training deep neural network with a limited amount of data to develop competitive deep learning model for aneurysm rupture prediction using only a limited amounts of data.
- 31
- MTMR-Net: Multi-task Deep Learning with Margin Ranking Loss for Lung Nodule Analysis31
A Multi-Task deep learning framework with a novel Margin Ranking loss (referred as MTMR-Net) for automated lung nodule analysis achieved competitive classification performance and more accurate attribute scoring over the state-of-the-arts.
- Integrating Artificial Intelligence and Augmented Reality in Robotic Surgery: An Initial dVRK Study Using a Surgical Education Scenario30
A novel system by seamlessly merging artificial intelligence module and augmented reality visualization to automatically generate the surgical guidance for robotic surgery education to prove its feasibility and potential as the next generation of robot-assisted surgery education solution.
- 29
- Towards fair decentralized benchmarking of healthcare AI algorithms with the Federated Tumor Segmentation (FeTS) challenge28
The authors describe the Federated Tumor Segmentation (FeTS) challenge for the decentralised benchmarking of FL algorithms and evaluation of Healthcare AI algorithm generalizability in real-world cancer imaging datasets.
- RGBManip: Monocular Image-based Robotic Manipulation through Active Object Pose Estimation28
This work proposes an image-only robotic manipulation framework that leverages an eye-on-hand monocular camera installed on the robot’s parallel gripper that enables the estimation of 6D object poses, which can be utilized for manipulation.
- 28
- Accurate instance segmentation of surgical instruments in robotic surgery: model refinement and cross-dataset evaluation27
This work presents an instance segmentation model based on refined Mask R-CNN for accurately segmenting the instruments as well as identifying their types, and provides guidance on how to allocate the annotation cost for surgeons while labelling a new dataset in practice.
- Interactive Navigation in Environments with Traversable Obstacles Using Large Language and Vision-Language Models26
The large language model (GPT-3.5) and the open-set Vision-language Model (Grounding DINO) are utilized to create an action-aware costmap to perform effective path planning without fine-tuning, allowing robots to navigate in environments with traversable obstacles.
- One to Many: Adaptive Instrument Segmentation via Meta Learning and Dynamic Online Adaptation in Robotic Surgical Video26
MDAL, a meta-learning based dynamic online adaptive learning scheme with a two-stage framework to fast adapt the model parameters on the first frame and partial subsequent frames while predicting the results, outperforms other state-of-the-art methods on two datasets (including a real-world RAS dataset).
- Edge-Aware Pyramidal Deformable Network for Unsupervised Registration of Brain MR Images26
A novel edge-aware pyramidal deformable network (referred as EPReg) is proposed for unsupervised volumetric registration to fully exploit the useful complementary information from the multi-level feature pyramids to predict multi-scale displacement fields.
- 26
- 25
- A review on organ deformation modeling approaches for reliable surgical navigation using augmented reality25
A comprehensive and technical-oriented overview of modeling methods for intraoperative organ deformation in augmented reality in surgery, and discusses the potential topics for future advancements is provided.
- 25
- Deep Angular Embedding and Feature Correlation Attention for Breast MRI Cancer Analysis25
The proposed COAM accomplished more accurate localization of the cancer center compared with other state-of-the-art weakly supervised localization method and localize possible cancer lesion by COrrelation Attention Map based on the learned features.
- 24
- 24
- 23
- 22
- FedDBL: Communication and Data Efficient Federated Deep-Broad Learning for Histopathological Tissue Classification22
A lightweight and universal FL framework, named federated deep-broad learning (FedDBL), to achieve superior classification performance with limited training samples and only one-round communication and can dramatically reduce data dependency and improve communication efficiency.
- Autonomous Intelligent Navigation for Flexible Endoscopy Using Monocular Depth Guidance and 3-D Shape Planning21
A novel data-driven framework with self-contained visual-shape fusion for autonomous intelligent navigation of flexible endoscopes requiring no priori knowledge of system models and global environments is presented.
- 21
- A Learning-Driven Framework with Spatial Optimization For Surgical Suture Thread Reconstruction and Autonomous Grasping Under Multiple Topologies and Environmental Noises21
A vision-based method to automate the suture thread grasping, which is a sub-task in surgical knot tying and an intermediate step between the stitching and looping manipulations, and a transfer-learning strategy first to fine-tune a pre-trained model by learning the information from large legacy surgical data and images obtained by the onsite equipment.
- Image-Level Harmonization of Multi-site Data Using Image-and-Spatial Transformer Networks21
It is demonstrated that ISTNs can be trained adversarially on a classification problem with simulated 2D data and it is shown that age regression and sex classification models trained on ISTN output improve generalization when training on data from one and testing on the other site.
- Joint affine and deformable three‐dimensional networks for brain MRI registration21
The proposed network provides accurate and robust volumetric registration without any pre-alignment requirement, which facilitates the end-to-end deformable registration.
- Adaptive Online Learning and Robust 3-D Shape Servoing of Continuum and Soft Robots in Unstructured Environments20
A versatile learning-based adaptive shape controller by leveraging proprioception of 3-D configuration from fiber Bragg grating (FBG) sensors, which can online estimate the unknown model of continuum robot against unexpected disturbances and exhibit an adaptive behavior to the unmodeled system without priori data exploration is proposed.
- 20
- Domain Adaptive Graph Infomax via Conditional Adversarial Networks20
This work proposes a graph infomax method that is domain adaptive, and results in real-world datasets validate the performance of the method in comparison with the state-of-the-art baselines.
- 19
- Accurate Grid Keypoint Learning for Efficient Video Prediction19
A new grid keypoint learning framework is designed, aiming at a robust and explainable intermediate keypoint representation for long-term efficient video prediction, and outperforms the state-of-the-art stochastic video prediction methods while saves more than 98% of computing resources.
- Kernel-based Unsupervised Embedding Alignment for Enhanced Visual Representation in Vision-language Models18
This work proposes a novel kernel-based method to align CLIP's visual representation with that of DINOv2, ensuring that the resulting embeddings maintain compatibility with text embeddings while enhancing perceptual capabilities.
- 18
- Future Frame Prediction for Robot-Assisted Surgery18
This paper proposes a ternary prior guided variational autoencoder (TPG-VAE) model for future frame prediction in robotic surgical video sequences, and is the first time that the future frames of dual arm robots are predicted considering their unique characteristics relative to general robotic videos.
- 18
- UC-NeRF: Uncertainty-Aware Conditional Neural Radiance Fields From Endoscopic Sparse Views17
This paper proposes uncertainty-aware conditional NeRF for novel view synthesis to tackle the severe shape-radiance ambiguity from sparse surgical views, and incorporates the multi-view uncertainty estimation to condition the neural radiance field for modeling the severe photometric inconsistencies adaptively.
- 17
- Imaging-Based Deep Graph Neural Networks for Survival Analysis in Early Stage Lung Cancer Using CT: A Multicenter Study17
The proposed GCN model outperformed all ML, Tumor-CNN, and TNM staging models and demonstrated the value of utilizing medical imaging graph structure data, resulting in a robust and effective model for the prediction of survival in early-stage lung cancer.
- 16
- 16
- Prothrombin Complex Concentrate vs Conservative Management in ICH Associated With Direct Oral Anticoagulants16
In this study, PCC treatment was not associated with improved functional outcome, hematoma expansion, or mortality among patients with DOAC-associated ICH.
- Robotic surgery remote mentoring via AR with 3D scene streaming and hand interaction16
A novel AR-based robotic surgery remote mentoring system with efficient 3D scene visualisation and natural hand interaction and holds great potential for clinical applications is proposed.
- Development of deep learning algorithms to discriminate giant cell tumors of bone from adjacent normal tissues by confocal Raman spectroscopy16
The results of this study suggest that confocal Raman spectroscopy accomplished by the LSTM network could non-destructively evaluate a tumor margin by its inherent biochemical specificity which may allow intraoperative assessment of the adequacy of tumor clearance.
- Anchor-guided online meta adaptation for fast one-Shot instrument segmentation from robotic surgical videos16
This paper designs an anchor-guided online adaptation to tackle the performance drop throughout a robotic surgical sequence, and achieves fast one-shot test time optimization by meta-learning a good model initialization and learning rates from source videos to avoid the laborious and handcrafted fine-tuning.
- 15
- 15
- 15
- Uncertainty Estimation for Safety-critical Scene Segmentation via Fine-grained Reward Maximization14
A novel fine-grained reward maximization (FGRM) framework, to address uncertainty estimation by directly utilizing an uncertainty metric related reward function with a reinforcement learning based model tuning algorithm, to benefit the model uncertainty estimation through direct optimization guidance for model calibration.
- A software program for automated compressive vertebral fracture detection on elderly women’s lateral chest radiograph: Ofeye 1.014
A user-friendly software program was developed for CVF detection on elderly women’s lateral chest radiographs which has an overall low false positivity rate, and for moderate and severe CVFs an acceptably low false negativity rate.
- 14
- A Two-Stage Approach for Automated Prostate Lesion Detection and Classification with Mask R-CNN and Weakly Supervised Deep Neural Network14
A novel two-stage method for fully-automated prostate lesion detection and classification, using input sequences of T2-weighted images, apparent diffusion coefficient (ADC) maps and high b-value diffusion-weighting images, is presented.
- 13
- OpenMEDLab: An Open-source Platform for Multi-modality Foundation Models in Medicine13
OpenMEDLab is presented, an open-source platform for multi-modality foundation models that encapsulates not only solutions of pioneering attempts in prompting and fine-tuning large language and vision models for frontline clinical and bioinformatic applications but also building domain-specific foundation models with large-scale multi-modal medical data.
- SegSTRONG-C: Segmenting Surgical Tools Robustly On Non-adversarial Generated Corruptions -- An EndoVis'24 Challenge13
The SegSTRONG-C challenge is introduced, calling for new paradigms that enhance universal robustness to unforeseen"corruptions" to facilitate richer applications in surgical data science and identifies practical approaches for enhancing model robustness.
- Holistic-Motion2D: Scalable Whole-body Human Motion Generation in 2D Space13
This paper presents Holistic-Motion2D, the first comprehensive and large-scale benchmark for 2D whole-body motion generation, which includes over 1M in-the-wild motion sequences, each paired with high-quality whole-body/partial pose annotations and textual descriptions.
- Unsupervised feature disentanglement for video retrieval in minimally invasive surgery13
A novel method of Unsupervised Disentanglement of Scene and Motion (UDSM) representations for minimally invasive surgery video retrieval within large databases, which has the potential to advance intelligent and efficient surgical teaching systems.
- 12
- 12
- End-to-End Learning of Deep Visuomotor Policy for Needle Picking12
This paper presents the first end- to-end learning method to train deep visuomotor policy for needle picking, and effectively demonstrated its superiority in generalization to unseen in-domain variations including needle variations and image disturbance, highlighting its robustness and versatility.
- Distilled Visual and Robot Kinematics Embeddings for Metric Depth Estimation in Monocular Scene Reconstruction12
This paper presents a novel framework that combines robot kinematics and monocular endoscope images with deep unsupervised learning into a single network for metric depth estimation and then achieves 3D reconstruction of complex anatomy.
- Multi-scale and Modality Dropout Learning for Intervertebral Disc Localization and Segmentation12
This paper designs a 3D fully convolutional network which takes multiple scales of images as input and merges these pathways at higher layers to jointly integrate multi-scale information and proposes a modality dropout strategy to alleviate the co-adaption issue during the training.
- 11
- 11
- 11
- 11
- 11
- Toward Modality- and Sampling-Universal Learning Strategies for Accelerating Cardiovascular Imaging: Summary of the CMRxRecon2024 Challenge10
Analysis of the best-performing solutions revealed that prompt-based adaptation and enhanced physics-driven consistency enabled strong cross-scenario performance and establish principles for generalizable reconstruction models and advance clinically translatable AI in cardiovascular imaging.
- 10
- 10
- Structured Gradient-Based Interpretations via Norm-Regularized Adversarial Training10
A duality relation is shown between the regularized norms of the adversarial perturbations and gradient-based maps, based on which adversarial training loss functions promoting sparsity and group-sparsity properties in simple gradient maps are designed.
- Self-Supervised Cyclic Diffeomorphic Mapping for Soft Tissue Deformation Recovery in Robotic Surgery Scenes10
This work introduces a novel self-supervised framework to recover tissue deformations from stereo surgical videos that integrates semantics, cross-frame motion flow, and long-range temporal dependencies to accurately represent tissue dynamics for deformation recovery.
- Simultaneous Estimation of Shape and Force along Highly Deformable Surgical Manipulators Using Sparse FBG Measurement10
A novel data-driven paradigm for simultaneous estimation of shape and force along highly deformable flexible robots by using sparse strain measurement from a single-core FBG fiber is proposed and three learning models are introduced by incorporating spatial strain encoders.
- Two-Stage Grasping: A New Bin Picking Framework for Small Objects10
Unlike traditional bin picking methods focusing on vision-based grasping planning using classic frameworks, the challenges of picking cluttered small objects can be solved by the proposed new framework with simple vision detection and planning.
- 10
- 10
- 10
- WonderVerse: Extendable 3D Scene Generation with Video Generative Models9
WonderVerse leverages the powerful world-level priors embedded within video generative foundation models to create highly immersive and geometrically coherent 3D environments, and proposes a new technique for controllable 3D scene extension to substantially increase the scale of the generated environments.
- Toward Reliable AR-Guided Surgical Navigation: Interactive Deformation Modeling With Data-Driven Biomechanics and Prompts9
A data-driven biomechanics algorithm that preserves FEM-level accuracy while improving computational efficiency is proposed and a novel human-in-the-loop mechanism is introduced into the deformation modeling process, enabling surgeons to interactively provide prompts to correct anatomical misalignments.
- Incorporating Clinical Guidelines Through Adapting Multi-modal Large Language Model for Prostate Cancer PI-RADS Scoring9
A novel approach is introduced that adapts a multi-modal large language model (MLLM) to incorporate PICG into PI-RADS scoring model without additional annotations and network parameters, which effectively improves the performance of current scoring networks.
- 9
- Efficient Physically-based Simulation of Soft Bodies in Embodied Environment for Surgical Robot9
An interactive physically based soft body simulation framework is developed and integrated to SurRoL that is able to simulate soft bodies efficiently with strong physical interpretability and plausible visual effects and enable the efficient simulation of surgical tasks involving soft tissue manipulation.
- Multi-objective Cross-task Learning via Goal-conditioned GPT-based Decision Transformers for Surgical Robot Task Automation9
A new learning-based framework by leveraging the strong reasoning capability of the GPT-based architecture to automate surgical robotic tasks by developing a goal-conditioned decision transformer to achieve sequential representations with goal-aware future indicators in order to enhance temporal reasoning.
- Rapid Endoscopic Diagnosis of Benign Ulcerative Colorectal Diseases With an Artificial Intelligence Contextual Framework9
A Multi-class Contextual Classification (MCC) model is developed that can differentiate among the mentioned UCDs and normal controls by incorporating the tissue object contexts surrounding the individual lesion region in a scene and spatial information from other endoscopic frames (video-level) into a unified framework.
- 9
- Efficient EndoNeRF reconstruction and its application for data-driven surgical simulation9
A novel NeRF-based reconstruction framework with an emphasis on simulation purposes is proposed that facilitates the efficient creation of high-quality surgical soft tissue 3D models and has the potential to benefit downstream clinical tasks, such as surgical education.
- 9
- Joint Optimization of Class-Specific Training- and Test-Time Data Augmentation in Segmentation9
This paper adopts a computationally efficient and data-efficient gradient-based meta-learning scheme to explicitly align the distribution of training and validation data which is used as a proxy for unseen test data to improve the current data augmentation strategies.
- 9
- 9
- Learning Deep Nets for Gravitational Dynamics With Unknown Disturbance Through Physical Knowledge Distillation: Initial Feasibility Study9
A novel framework characterized by both high data efficiency and disturbance-adapting capability is proposed to address the problem of modeling gravitational dynamics using deep nets in feedforward gravity compensation control for high-DOF master manipulators with unknown disturbance.
- Unsupervised Domain Adaptation of ConvNets for Medical Image Segmentation via Adversarial Learning9
This chapter presents unsupervised domain adaptation methods using adversarial learning, to generalize the ConvNets for medical image segmentation tasks, and presents solutions from two different perspectives, i.e., feature- level adaptation and pixel-level adaptation.
- Extreme cardiac MRI analysis under respiratory motion: Results of the CMRxMotion challenge8
A comprehensive overview of the challenge design and dataset is presented, the evaluation results for the top-performing methods are reported, and the impact of motion artifacts on five clinically relevant biomarkers are investigated.
- 8
- 8
- 8
- Pseudo-label Guided Cross-video Pixel Contrast for Robotic Surgical Scene Segmentation with Limited Annotations8
This work proposes PGV-CL, a novel pseudo-label guided cross-video contrast learning method that effectively leverages unlabeled data for a trusty and global model regularization that produces more discriminative feature representation in surgical scene segmentation.
- Rapid Model Transfer for Medical Image Segmentation Via Iterative Human-in-the-Loop Update: from Labelled Public to Unlabelled Clinical Datasets for Multi-Organ Segmentation in CT8
A novel and generic human-in-the-loop scheme for efficiently transferring a segmentation model from a small-scale labelled dataset to a larger-scale unlabelled dataset for multi-organ segmentation in CT.
- Learning dissection trajectories from expert surgical videos via imitation learning with equivariant diffusion7
This paper introduces imitation learning for the critical task of predicting dissection trajectories from expert video demonstrations with a novel Implicit Diffusion Policy with Equivariant Representations for Imitation Learning (iDPOE) to address variability.
- Feasibility of real-time artificial intelligence-assisted anatomical structure recognition during endoscopic submucosal dissection7
In this proof-of-concept study, feasibility of a real-time AI algorithm for anatomical structure delineation and orientation support during ESD proved safe and appropriate for routine procedures in humans.
- 7
- Open-Vocabulary Object Detection with Meta Prompt Representation and Instance Contrastive Optimization7
This work proposes a novel framework with Meta prompt and Instance Contrastive learning (MIC) schemes that outperforms previous SOTA methods trained with these complex techniques on LVIS and shows great generalization ability on novel classes.
- 7
- 7
- 7
- Automatic lesion detection with three-dimensional convolutional neural networks7
This chapter presents a 3-D convolutional neural network–based cascaded framework to effectively conduct the automatic lesion detection task and describes in detail the principle of 3- D convolution kernels, the construction of3-D networks, the merit of fully Convolutional architectures, and the specifics of a two-step workflow.
- Scale-aware monocular reconstruction via robot kinematics and visual data in neural radiance fields6
Inspired by recently proposed neural radiance fields (NeRF), this work presents a novel pipeline, KV-EndoNeRF, which explores limited multi-modal data for surgical scene reconstruction with absolute scale, and excels in learning an absolute scale from robot kinematics.
- 6
- 6
- SeamlessNeRF: Stitching Part NeRFs with Gradient Propagation6
This work proposes SeamlessNeRF, a novel approach for seamless appearance blending of multiple NeRFs, and is the first work that introduces gradient-guided appearance editing to radiance fields, offering solutions for seamless stitching of 3D objects represented in Ne RFs.
- Shape Mask Generator: Learning to Refine Shape Priors for Segmenting Overlapping Cervical Cytoplasms6
A conceptually simple and effective technique, called shape mask generator, for segmenting overlapping cytoplasms, which model shape priors from shape templates and feed them to the shape mask generators that generates a shape mask for the cytoplasm as the segmentation result.
- Deep Cascaded Networks for Sparsely Distributed Object Detection from Medical Images6
This chapter introduces a two-stage cascaded deep learning framework to detect sparsely distributed objects that provide clinical significance with both high efficiency and accuracy and corroborates the importance of volumetric feature representations forvolumetric imaging modalities by exploiting 3D convolutional neural networks.
- A retrieval-augmented framework enabling VLM spatial awareness for object-centric robot manipulation5
Retrieval-Augmented Manipulation (RAM) is introduced, an object-centric framework that endows general-purpose vision foundation models with the spatial reasoning necessary for robust manipulation and represents a critical step toward developing more physically intelligent and general-purpose robotic systems.
- 5
- Illuminating the unseen: Advancing MRI domain generalization through causality5
This work proposes a novel approach enforcing mechanism-level invariance, termed GenCA-MRI, which aligns intrinsic causal relationships within MRI data and develops a computational strategy that significantly reduces the complexity of causal alignment, ensuring its feasibility for real-world applications.
- 5
- 5
- Ada-Tracker: Soft Tissue Tracking via Inter-Frame and Adaptive-template Matching5
This work exploits optical flow to naturally capture the pixel-wise tissue deformations and adaptively correct the tracked template and proposes an adaptive-template matching method, which updates the tracked template based on the reliability of the estimates.
- 5
- PEg TRAnsfer Workflow recognition challenge report: Do multimodal data improve recognition?5
The improvement of surgical workflow recognition methods using multiple modalities compared with unimodal methods was significant for all teams, however, the longer execution time required for video/kinematic-based methods(compared to only kinematic- based methods) must be considered.
- Adaptive Negative Evidential Deep Learning for Open-set Semi-supervised Learning5
This paper first introduces evidential deep learning (EDL) as an outlier detector to quantify different types of uncertainty, and design different uncertainty metrics for self-training and inference, and proposes a novel adaptive negative optimization strategy.
- Automatic Multiparametric Magnetic Resonance Imaging‐Based Prostate Lesions Assessment with Unsupervised Domain Adaptation5
A coarse mask‐guided deep domain adaptation network (CMD2A‐Net) is proposed to develop a fully automated framework for prostate lesion detection and classification (PLDC).
- Author Correction: Federated learning enables big data for rare cancer boundary detection5
This exhibition aims to showcase the talent in the rapidly changing field of mobilemented reality and aims to provide a platform for up-and-coming artists to showcase their talent.
- Development and Evaluation of Machine Learning in Whole-Body Magnetic Resonance Imaging for Detecting Metastases in Patients With Lung or Colon Cancer5
A machine learning algorithm was developed to improve radiologists' sensitivity and specificity for metastasis detection and reduce reading times and there was no evidence of a significant difference in per-patient and specificity.
- 5
- pFedFair: Towards Optimal Group Fairness-Accuracy Trade-off in Heterogeneous Federated Learning4
This work proposes the Personalized Federated Learning for Client-Level Group Fairness (pFedFair) framework, and extends the application of pFedFair to computer vision settings, where it is numerically shown that pFedFair achieves an optimal group fairness-accuracy trade-off in heterogeneous FL settings.
- Small but Strong: Lightweight Architecture Improves Lung Nodule Detection and Segmentation4
This study proposes a lightweight architecture for lung nodule detection and segmentation, alleviating the overfitting caused by the data scarcity, and outperforms the state‐of‐the‐art methods with a small model size.
- 4
- On-the-Fly Improving Segment Anything for Medical Image Segmentation Using Auxiliary Online Learning4
- 4
- 4
- 4
- SQA-SAM: Segmentation Quality Assessment for Medical Images Utilizing the Segment Anything Model4
A novel SQA method is proposed, called SQA-SAM, which exploits SAM to enhance the accuracy of quality assessment for medical image segmentation and develops a score measure for such alignment.
- Author Correction: Federated deep learning for detecting COVID-19 lung abnormalities in CT: a privacy-preserving multinational validation study4
The federated learning algorithms are explored to develop a privacy-preserving AI model for COVID-19 medical image diagnosis with good generalization capability on unseen multinational datasets.
- 4
- Domain Adaptive Robotic Gesture Recognition with Unsupervised Kinematic-Visual Data Alignment4
A novel unsupervised domain adaptation framework which can simultaneously transfer multi-modality knowledge, i.e., both kinematic and visual data, from simulator to real robot, is proposed and recovers the performance with great improvement gains.
- Domain Knowledge Driven Multi-modal Segmentation of Anatomical Brain Barriers to Cancer Spread4
A multi-modal segmentation method largely driven by domain knowledge is explored, which applies 3D U-Net as the backbone model and employs a label merging strategy for the symmetrical structures which have both left and right labels, to highlight the structural information regardless of the locations.
- 4
- 3
- 3
- 3
- 3
- 3
- 3
- Efficient Transferability Assessment for Selection of Pre-trained Detectors3
Experimental results demonstrate that the method outperforms other state-of-the-art approaches in assessing transferability under different target domains while efficiently reducing wall-clock time 32× and requires a mere 5.2% memory footprint compared to brute-force fine-tuning of all pretrained detectors.
- 3
- Visual-Kinematics Graph Learning for Procedure-Agnostic Instrument Tip Segmentation in Robotic Surgeries3
This work proposes a novel visual-kinematics graph learning framework to accurately segment the instrument tip given various surgical procedures, and significantly outperformed current image-based state-of-the-art approaches.
- 3
- Can Generalist Vision Language Models (VLMs) Rival Specialist Medical VLMs? Benchmarking and Strategic Insights2
It is suggested that leveraging common VLMs with fine-tuning offers a scalable and cost-effective alternative to developing large-scale medical VLMs, providing crucial insights for future research in the medical imaging field.
- Dynamic key vascular anatomy dataset for D2 lymph node dissection during laparoscopic gastric cancer surgery2
The LapGC-KVAD-30 uniquely contains 5303 frames that showcase the dynamic process of key vessels from initial appearance to full exposure (or ligation), providing essential information for effective and safe LND.
- HiEndo: harnessing large-scale data for generating high-resolution laparoscopy videos under a two-stage framework2
This work proposes an innovative two-stage video generation architecture HiEndo for generating high-resolution gastrointestinal laparoscopy videos with high fidelity, which can meet the real-world clinical usage.
- UCAgents: Unidirectional Convergence for Visual Evidence Anchored Multi-Agent Medical Decision-Making2
Inspired by clinical workflows, UCAgents forbids position changes and limits agent interactions to targeted evidence verification, suppressing rhetorical drift while amplifying visual signal extraction in a hierarchical multi-agent framework enforcing unidirectional convergence through structured evidence auditing.
- Markerless Tracking of Robotic Surgical Instruments With Head Mounted Display for Augmented Reality Applications2
A novel markerless tracking method for robotic instruments using a HoloLens 2 HMD is proposed that suggests that the markerless method is an acceptable substitute to marker‐based tracking for this augmented reality application, while avoiding workflow issues with sterilizing and attaching markers.
- Hepatic arterial infusion chemotherapy combined with bevacizumab plus a PD-1 inhibitor for gallbladder cancer with hepatic oligometastasis: a real-world study2
The combination of HAIC, a PD-1 inhibitor, and bevacizumab shows potential for advanced GBC with hepatic oligometastasis and the therapeutic response of hepatic metastasis had a greater influence on prognosis than that of primary gallbladder lesions.
- A Super-pixel-based Approach to the Stable Interpretation of Neural Networks2
This work proposes a novel pixel partitioning strategy based on super-pixels which cluster pixels into groups that align well with the semantic meaning of the images and demonstrates that the grouping of pixels reduces the variance of the saliency map and improves the generalization behavior of the interpretation method.
- 2
- 2
- 2
- Patch-Based DCNN Method for CBCT Image Enhancement2
A deep convolutional neural network (DCNN) based method for head and neck (HN) CBCT enhancement, where the network learns to transform the low-quality CBCT images to high-quality images, showing promising results for fast CBCT image enhancement.
- 2
- Real-Time Monocular 2-D and 3-D Perception of Endoluminal Scenes for Controlling Flexible Robotic Endoscopic Instruments1
An integrated visual perception platform for a continuum robotic system in endoluminal surgery to leverage monocular endoscopic image-based perception algorithms to accurately identify the position and orientation of flexible instruments and measure their distances from surrounding tissues is presented.
- 1
- 1
- Machine Learning Predictive Models for Survival in Colorectal Cancer Patients With Diabetes: A Cohort Study of 10 749 Subjects1
To identify the risk factors for the survival of colorectal cancer patients with type 2 diabetes mellitus with type 2 diabetes mellitus (T2DM), compare the predictive performance of models based on different algorithms, and develop a risk score system to predict the survival risk of the target population.
- 1
- 1
- 1
- Machine Learning Predictive Models for Survival in Gastric Cancer Patients with Diabetes: A Population-Based Cohort Study1
The findings indicate that older age at cancer diagnosis, longer duration of T2DM, higher body mass index (BMI), central obesity, lower levels of high-density lipoprotein cholesterol, and reduced serum potassium are associated with poorer prognosis for gastric cancer in patients with T2DM.
- 1
- 1
- 1
- 2244 Effect of long-term care insurance in a pilot city of China: health benefits among 12,930 disabled older adults1
This research marks the first comprehensive exploration of the potential health benefits associated with the implementation of LTCI, providing valuable perspectives that can inform policy making and enhance the development of robust long-term care systems in developing countries.
- 1
- Virtual Reality-Based Infrared Pupillometry (VIP) for Long-COVID1
Combining statistical methods to identify specific pupillometric features and ML algorithms to analyse the performance further enhance the performance of VIP as a non-intrusive, low-cost, portable and objective method to detect and monitor long COVID.
- Improving Segment Anything on the Fly: Auxiliary Online Learning and Adaptive Fusion for Medical Image Segmentation1
A novel approach that leverages the advantages of online machine learning to enhance Segment Anything (SA) during test time and validates a new, practical, and effective approach for enhancing SA on downstream segmentation tasks (e.g., medical image segmentation).
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- MotionDreamer: Generating Continuous Intermediate Motions for Complex Motion Interpolation–
MotionDreamer is a novel framework that instills effective motion priors for all intermediate frames to video generators by regularizing diffusion training with ground truth continuous intermediate motions and collects the first open-source large-scale dataset, VFIVid, for complex motion interpolation.
- UCMamba: Visual State Space Model for Ulcerative Colitis Severity Scoring with Spiral Scan and Sequential Contrastive Learning–
UCMamba is proposed, a Visual State Space Model tailored to UC severity scoring that adopts a new sequence-aware paradigm for selecting positive and negative samples, and incorporates severity score distances as an adaptive margin, which enables more discriminative and clinically meaningful feature learning, thereby improving prediction performance.
- Enabling highly accelerated multi-shot IVIM-DWI with a Data-sharing-guided framework: Joint image reconstruction and biomarker estimation–
The effectiveness of the proposed data-sharing-guided framework for highly accelerated multi-shot IVIM-DWI has the potential to improve quantitative IVIM-DWI assessment in cerebrovascular and neurological diseases.
- –
- –
- 3D reconstruction of colored objects via Gamma correction and piecewise normalization–
Experimental results show that the method proposed in this paper can not only achieve accurate measurement of monochromatic objects but also measure colorful objects.
- –
- –
- –
- –
- –
- GG01-21 DEVELOPING AND VALIDATING AN ADVANCED ARTIFICIAL INTELLIGENCE (AI) SEGMENTATION MODEL TO ASSIST UROLOGISTS IN MRI-ULTRASOUND FUSION BIOPSY–
The model could segment the prostate, TZ and PZ automatically and accurately and shows its potential in assisting urologists to perform MRI-ultrasound fusion biopsy, while further validation and clinical trials are needed.
- –
- –
- –
- –
- –
- –
- –
- –
- –
- Holistic Consistency for Subject-Level Segmentation Quality Assessment in Medical Image Segmentation–
- –
- –
- Predicting Knee Osteoarthritis Progression with an Interpretable Deep Learning Approach on MRIs: Data from the Osteoarthritis Initiative–
It is shown that the deep learning framework performed well on discrimination of progressors, especially at 24th months, and that the infrapatellar fat pad plays an important role in predicting progression.
- –
- –
- Can infrapatellar fat pad predict the incidence of knee osteoarthritis by using deep learning based on MRI? Data from osteoarthritis initiative–
A deep learning model is trained to predict iROA with auto-segmented IPFP and the results reveal that IPFP alteration can predict i ROA independently comparably to the whole knee MRI at one year before iRO a.
- –
- –
- –
- Special issue on 2020 augmented environments for computer-assisted interventions (AE-CAI): guest editors’ foreword–
This Special Issue of Taylor & Francis’s Computer Methods in Biomechanics and Biomedical Engineering: Imaging & Visualisation (TCIV) journal dedicated to the joint edition of the Augment...
- –
- –
- –
- –
- Deep multilevel contextual networks for biomedical image segmentation–
A deep contextual network is proposed here by leveraging multilevel contextual information from the deep hierarchical structure to achieve better segmentation performance and can outperform the state-of-the-art methods by a large margin with respect to different evaluation measures.
- –
- –
Publication data from OpenAlex, with missing venues and authors filled in from Crossref; citation counts are the higher of OpenAlex and Semantic Scholar, last synced 2026-10-12. One-sentence summaries under some papers are written by Semantic Scholar’s model. Citation counts may be lower than on Google Scholar, which indexes more sources.
Report an error
Wrong papers, two people merged into one, or a profile that should not be here? Tell us and we will fix or hide it.