Showing papers in "Medical Image Analysis in 2022"
••
TL;DR: An overview of explainable artificial intelligence (XAI) used in deep learning-based medical image analysis can be found in this article , where a framework of XAI criteria is introduced to classify deep learning based medical image classification methods.
94 citations
••
TL;DR: The HEAD and neCK TumOR (HECKTOR) challenge as discussed by the authors was the first one focusing on lesion segmentation in combined FDG-PET and CT image modalities.
85 citations
••
TL;DR: A comprehensive overview of applying deep learning methods in various medical image analysis tasks can be found in this paper , where the authors highlight the latest progress and contributions of state-of-the-art unsupervised and semi-supervised deep learning in medical image classification, segmentation, detection and image registration.
78 citations
••
TL;DR: Zhang et al. as discussed by the authors proposed a feature adaptive transformer network (FAT-Net) which integrates an extra transformer branch to capture long-range dependencies and global context information, and employed a memory-efficient decoder and a feature adaptation module to enhance the feature fusion between the adjacent-level features by activating the effective channels and restraining the irrelevant background noise.
74 citations
••
TL;DR: In this article , a federated approach for weakly supervised computational pathology on gigapixel whole-slide images was proposed, which can effectively develop accurate weak-supervised deep learning models from distributed data silos without direct data sharing and its associated complexities.
57 citations
••
TL;DR: Surgical Data Science (SDS) is a new research field that aims to improve the quality of interventional healthcare through the capture, organization, analysis and modeling of data as mentioned in this paper .
56 citations
••
TL;DR: TransMorph as mentioned in this paper is a hybrid Transformer-ConvNet model for volumetric medical image registration, which combines the topology-preserving deformations of moving and fixed images.
44 citations
••
TL;DR: ResGANet as mentioned in this paper proposes a modular group attention block that can capture feature dependencies in medical images in two independent dimensions: channel and space, by stacking these group attention blocks in ResNet-style.
43 citations
••
TL;DR: Kather et al. as mentioned in this paper compared the performance of weakly-supervised and multiple-instance learning (MIL) based approaches for histopathological slide subtyping of renal cell carcinoma (RCC) in colorectal, gastric and bladder cancer.
43 citations
••
TL;DR: In this paper , the latent representation of vector quantised variational autoencoders with an ensemble of autoregressive transformers is combined to enable unsupervised anomaly detection and segmentation defined by deviation from healthy brain imaging data.
42 citations
••
TL;DR: The first Fast and Low GPU memory Abdominal ORgan sEgmentation (FLARE) challenge as discussed by the authors was organized to comprehensively benchmark abdominal organ segmentation methods, and the winning method surpassed the existing state-of-the-art method, achieving a 19× faster inference speed and reducing the GPU memory consumption by 60% with comparable accuracy.
••
TL;DR: Wang et al. as mentioned in this paper proposed a multi-task ViT that leverages low-level CXR feature corpus obtained from a backbone network that extracts common CXRs findings.
••
TL;DR: Wang et al. as discussed by the authors proposed a multi-task ViT that leverages low-level CXR feature corpus obtained from a backbone network that extracts common CXRs findings.
••
TL;DR: Li et al. as discussed by the authors proposed a multi-scale residual encoding and decoding network (Ms RED) for skin lesion segmentation, which is able to accurately and reliably segment a variety of lesions with efficiency.
••
TL;DR: Wang et al. as mentioned in this paper presented a simple yet efficient consistency regularization approach for semi-supervised medical image segmentation, called Uncertainty Rectified Pyramid Consistency (URPC).
••
TL;DR: In this article, a hierarchical graph neural network is proposed to operate on the hierarchical entity-graph and map the tissue structure to tissue functionality, treating the tissue as a hierarchical composition of multiple types of histological entities from fine to coarse level.
••
TL;DR: Li et al. as mentioned in this paper proposed a multi-scale residual encoding and decoding network (Ms RED) for skin lesion segmentation, which is able to accurately and reliably segment a variety of lesions with efficiency.
••
TL;DR: Zhang et al. as discussed by the authors introduced two auxiliary tasks, i.e., a foreground and background reconstruction task for capturing semantic information and a signed distance field (SDF) prediction task for imposing shape constraint, and explore the mutual promotion effect between the two auxiliary and the segmentation tasks based on mean teacher architecture.
••
TL;DR: The International Skin Imaging Collaboration (ISIC) dataset has become a leading repository for researchers in machine learning for medical image analysis, especially in the field of skin cancer detection and malignancy assessment as mentioned in this paper .
••
TL;DR: The MICCAI MIDOG 2021 challenge as mentioned in this paper was the first attempt to derive scanner-agnostic mitosis detection algorithms, which used a training set of 200 cases, split across four scanning systems.
••
TL;DR: Wang et al. as discussed by the authors proposed a semi-relevant contrastive learning (SRCL) strategy to align multiple positive instances with similar visual concepts, which increases the diversity of positives and then results in more informative representations.
••
TL;DR: The International Skin Imaging Collaboration (ISIC) dataset has become a leading repository for researchers in machine learning for medical image analysis, especially in the field of skin cancer detection and malignancy assessment as discussed by the authors.
••
TL;DR: In this paper , a hierarchical graph neural network is proposed to operate on the hierarchical entity-graph and map the tissue structure to tissue functionality, treating the tissue as a hierarchical composition of multiple types of histological entities from fine to coarse level.
••
TL;DR: In this paper , a Fully Transformer Network (FTN) is proposed to learn long-range contextual information for skin lesion analysis, which has linear computational complexity as it introduces a spatial pyramid pooling (SPP) module into multi-head attention.
••
TL;DR: Zhang et al. as mentioned in this paper proposed a boundary-aware context neural network (BA-Net) for 2D medical image segmentation to capture richer context and preserve fine spatial information, which incorporates encoder-decoder architecture.
••
TL;DR: CrossMoDA as mentioned in this paper is the first large and multi-class benchmark for unsupervised cross-modality domain adaptation in medical image segmentation, which aims to segment two key brain structures involved in the follow-up and treatment planning of vestibular schwannoma (VS): the VS and the cochleas.
••
TL;DR: Wang et al. as mentioned in this paper proposed a retrieval with clustering-guided contrastive learning (RetCCL) framework for robust and accurate WSI-level image retrieval, which integrates a novel self-supervised feature learning method and a global ranking and aggregation algorithm for much improved performance.
••
TL;DR: In this paper , a self-supervised pretext task was proposed to learn a powerful supervisory signal for unsupervised representation learning, and a new teacher-student semisupervised consistency paradigm was introduced to learn to effectively transfer the pretrained representations to downstream tasks based on prediction consistency with the task-specific unlabeled data.
••
TL;DR: In this paper, a self-supervised pretext task was proposed to learn a powerful supervisory signal for unsupervised representation learning, and a new teacher-student semisupervised consistency paradigm was introduced to learn to effectively transfer the pretrained representations to downstream tasks based on prediction consistency with the task-specific unlabeled data.
••
TL;DR: Wang et al. as mentioned in this paper developed a hierarchical deep learning framework for BM nucleated differential count (NDC) analysis on whole-slide images (WSIs), which can replace traditional manual counting relying on oil-immersion 100x objective lens with a total 1000x magnification.