Related references
Note: Only part of the references are listed.
Article
Engineering, Biomedical
David Ahmedt-Aristizabal et al.
Summary: This article introduces the application of graph analytics in digital pathology, including entity-graph construction and graph architectures, and their success in tumor localization and classification, tumor invasion and staging, image retrieval, and survival prediction. The methods are systematically organized based on the graph representation of the input image, scale, and organ, and the limitations of existing techniques and potential future research directions are outlined.
COMPUTERIZED MEDICAL IMAGING AND GRAPHICS
(2022)
Article
Computer Science, Artificial Intelligence
Pushpak Pati et al.
Summary: The accurate diagnosis, prognosis, and therapy response predictions for cancer patients rely heavily on the phenotype and distribution of histological entities in tissue specimens. Various methods have been developed to represent tissue structures using cell-graphs, leveraging graph theory and machine learning. This study proposes a novel hierarchical entity graph representation for tissue specimens and introduces a hierarchical graph neural network to map tissue structure to functionality. Through evaluation with the BRACS dataset, the proposed method demonstrates superior classification results compared to alternative methods and individual pathologists.
MEDICAL IMAGE ANALYSIS
(2022)
Article
Computer Science, Interdisciplinary Applications
Yi Zheng et al.
IEEE TRANSACTIONS ON MEDICAL IMAGING
(2022)
Article
Computer Science, Interdisciplinary Applications
Zhihua Wang et al.
Summary: In this paper, a transformer-guided framework is proposed to predict lymph node metastasis of papillary thyroid carcinoma from whole slide histopathological images (WSIs). The framework incorporates a lightweight feature extractor, clustering-based instance selection, and a transformer-MIL module to improve accuracy. Additionally, an attention-based mutual knowledge distillation paradigm is utilized to enhance the training of the model. Experimental results demonstrate that the proposed approach outperforms state-of-the-art methods in terms of AUC.
IEEE TRANSACTIONS ON MEDICAL IMAGING
(2022)
Proceedings Paper
Computer Science, Artificial Intelligence
Richard J. Chen et al.
Summary: A new ViT architecture called HIPT is introduced to leverage the hierarchical structure inherent in gigapixel WSIs, with impressive performance on 9 slide-level tasks.
2022 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR 2022)
(2022)
Proceedings Paper
Computer Science, Artificial Intelligence
Kevin Thandiackal et al.
Summary: This paper introduces a method called ZoomMIL, which achieves high performance in classifying gigapixel-sized Whole-Slide Images (WSIs) in digital pathology. The method builds WSI representations by aggregating tissue-context information from multiple magnifications, significantly reducing computational demands.
COMPUTER VISION, ECCV 2022, PT XXI
(2022)
Proceedings Paper
Computer Science, Artificial Intelligence
Hongrun Zhang et al.
Summary: This paper investigates the multiple instance learning problem in the classification of histopathology whole slide images, and proposes a double-tier MIL framework for small sample cohorts. It also introduces the concept of pseudo-bags and utilizes attention-based MIL framework to calculate instance probability. The proposed method outperforms other approaches on multiple datasets.
2022 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR 2022)
(2022)
Proceedings Paper
Computer Science, Artificial Intelligence
Haowei Zhu et al.
Summary: In this work, the authors propose a dual cross-attention learning algorithm to extend self-attention modules for fine-grained object recognition. They introduce global-local cross-attention and pairwise cross-attention to enhance interactions between global images, local high-response regions, and image pairs. The algorithm reduces misleading attentions and improves the recognition performance on various tasks.
2022 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR 2022)
(2022)
Proceedings Paper
Computer Science, Artificial Intelligence
Yu Zhao et al.
Summary: This article introduces a Multiple Instance Learning (MIL) method called SETMIL, which is based on spatial encoding transformers. It effectively addresses pathological image analysis tasks and provides multi-scale context information, achieving superior performance.
MEDICAL IMAGE COMPUTING AND COMPUTER ASSISTED INTERVENTION, MICCAI 2022, PT II
(2022)
Proceedings Paper
Computer Science, Artificial Intelligence
Zhilong Lv et al.
Summary: This paper proposes a joint region-attention and multi-scale transformer (RAMST) network for microsatellite instability detection from whole slide images in gastrointestinal cancer. Compared to existing MSI detection methods, RAMST achieves the best performance and provides an effective feature representation learning method for WSI-label tasks.
MEDICAL IMAGE COMPUTING AND COMPUTER ASSISTED INTERVENTION, MICCAI 2022, PT II
(2022)
Review
Medicine, Research & Experimental
Miao Cui et al.
Summary: Computational pathology, driven by data processing and clinical informatics, offers innovative solutions for patient care. However, challenges such as data integration, hardware limitations, and talent development need to be addressed for the field to advance further.
LABORATORY INVESTIGATION
(2021)
Article
Engineering, Biomedical
Ming Y. Lu et al.
Summary: The CLAM method utilizes attention-based learning to identify subregions with high diagnostic value for accurate classification of whole-slide images. It can localize well-known morphological features without the need for spatial labels, outperforming standard weakly supervised classification algorithms, and adapt to independent test cohorts, smartphone microscopy, and varying tissue content.
NATURE BIOMEDICAL ENGINEERING
(2021)
Proceedings Paper
Computer Science, Artificial Intelligence
Andrey Zhmoginov et al.
Summary: This new method utilizes the Information Bottleneck principle to generate image attention masks in a semi-supervised setting, producing Boolean masks that conceal information in masked-out pixels effectively. It has been demonstrated to successfully attend to features defining the image class in synthetic datasets based on MNIST, CIFAR10, and SVHN datasets.
MACHINE LEARNING AND KNOWLEDGE DISCOVERY IN DATABASES, ECML PKDD 2020, PT III
(2021)
Proceedings Paper
Acoustics
Hang Li et al.
Summary: The article introduces a novel embedded-space MIL model, based on deformable transformer (DT) architecture and convolutional layers, termed DT-MIL, which outperforms other MIL models by generating the bag representation in a fully trainable way, representing the bag with a high-level and nonlinear combination of all instances, and encoding the position relationship and context information during bag embedding phase.
MEDICAL IMAGE COMPUTING AND COMPUTER ASSISTED INTERVENTION - MICCAI 2021, PT VIII
(2021)
Article
Computer Science, Interdisciplinary Applications
Wei Shao et al.
IEEE TRANSACTIONS ON MEDICAL IMAGING
(2020)
Article
Multidisciplinary Sciences
Osamu Iizuka et al.
SCIENTIFIC REPORTS
(2020)
Proceedings Paper
Computer Science, Artificial Intelligence
Jiawen Yao et al.
MEDICAL IMAGE COMPUTING AND COMPUTER ASSISTED INTERVENTION - MICCAI 2019, PT I
(2019)
Proceedings Paper
Computer Science, Theory & Methods
Aicha BenTaieb et al.
MEDICAL IMAGE COMPUTING AND COMPUTER ASSISTED INTERVENTION - MICCAI 2018, PT II
(2018)
Article
Computer Science, Information Systems
Luis Gonzalo Sanchez Giraldo et al.
IEEE TRANSACTIONS ON INFORMATION THEORY
(2015)
Article
Computer Science, Artificial Intelligence
Chang Xu et al.
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
(2014)
