1
|
Wiedenmann M, Barch M, Chang PS, Giltnane J, Risom T, Zijlstra A. An Immunofluorescence-Guided Segmentation Model in Hematoxylin and Eosin Images Is Enabled by Tissue Artifact Correction Using a Cycle-Consistent Generative Adversarial Network. Mod Pathol 2024; 37:100591. [PMID: 39147031 DOI: 10.1016/j.modpat.2024.100591] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 04/15/2024] [Accepted: 08/01/2024] [Indexed: 08/17/2024]
Abstract
Despite recent advances, the adoption of computer vision methods into clinical and commercial applications has been hampered by the limited availability of accurate ground truth tissue annotations required to train robust supervised models. Generating such ground truth can be accelerated by annotating tissue molecularly using immunofluorescence (IF) staining and mapping these annotations to a post-IF hematoxylin and eosin (H&E) (terminal H&E) stain. Mapping the annotations between IF and terminal H&E increases both the scale and accuracy by which ground truth could be generated. However, discrepancies between terminal H&E and conventional H&E caused by IF tissue processing have limited this implementation. We sought to overcome this challenge and achieve compatibility between these parallel modalities using synthetic image generation, in which a cycle-consistent generative adversarial network was applied to transfer the appearance of conventional H&E such that it emulates terminal H&E. These synthetic emulations allowed us to train a deep learning model for the segmentation of epithelium in terminal H&E that could be validated against the IF staining of epithelial-based cytokeratins. The combination of this segmentation model with the cycle-consistent generative adversarial network stain transfer model enabled performative epithelium segmentation in conventional H&E images. The approach demonstrates that the training of accurate segmentation models for the breadth of conventional H&E data can be executed free of human expert annotations by leveraging molecular annotation strategies such as IF, so long as the tissue impacts of the molecular annotation protocol are captured by generative models that can be deployed prior to the segmentation process.
Collapse
Affiliation(s)
- Marcel Wiedenmann
- Department of Computer and Information Science, University of Konstanz, Konstanz, Germany
| | - Mariya Barch
- Department of Research Pathology, Genentech Inc, South San Francisco, California
| | - Patrick S Chang
- Department of Research Pathology, Genentech Inc, South San Francisco, California
| | - Jennifer Giltnane
- Department of Research Pathology, Genentech Inc, South San Francisco, California
| | - Tyler Risom
- Department of Research Pathology, Genentech Inc, South San Francisco, California.
| | - Andries Zijlstra
- Department of Research Pathology, Genentech Inc, South San Francisco, California; Department of Pathology, Microbiology and Immunology, Vanderbilt University Medical Center, Nashville, Tennessee
| |
Collapse
|
2
|
Sierra-Jerez F, Martinez F. A non-aligned translation with a neoplastic classifier regularization to include vascular NBI patterns in standard colonoscopies. Comput Biol Med 2024; 170:108008. [PMID: 38277922 DOI: 10.1016/j.compbiomed.2024.108008] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 08/25/2023] [Revised: 12/21/2023] [Accepted: 01/13/2024] [Indexed: 01/28/2024]
Abstract
Polyp vascular patterns are key to categorizing colorectal cancer malignancy. These patterns are typically observed in situ from specialized narrow-band images (NBI). Nonetheless, such vascular characterization is lost from standard colonoscopies (the primary attention mechanism). Besides, even for NBI observations, the categorization remains biased for expert observations, reporting errors in classification from 59.5% to 84.2%. This work introduces an end-to-end computational strategy to enhance in situ standard colonoscopy observations, including vascular patterns typically observed from NBI mechanisms. These retrieved synthetic images are achieved by adjusting a deep representation under a non-aligned translation task from optical colonoscopy (OC) to NBI. The introduced scheme includes an architecture to discriminate enhanced neoplastic patterns achieving a remarkable separation into the embedding representation. The proposed approach was validated in a public dataset with a total of 76 sequences, including standard optical sequences and the respective NBI observations. The enhanced optical sequences were automatically classified among adenomas and hyperplastic samples achieving an F1-score of 0.86%. To measure the sensibility capability of the proposed approach, serrated samples were projected to the trained architecture. In this experiment, statistical differences from three classes with a ρ-value <0.05 were reported, following a Mann-Whitney U test. This work showed remarkable polyp discrimination results in enhancing OC sequences regarding typical NBI patterns. This method also learns polyp class distributions under the unpaired criteria (close to real practice), with the capability to separate serrated samples from adenomas and hyperplastic ones.
Collapse
Affiliation(s)
- Franklin Sierra-Jerez
- Biomedical Imaging, Vision and Learning Laboratory (BIVL(2)ab), Universidad Industrial de Santander (UIS), Colombia
| | - Fabio Martinez
- Biomedical Imaging, Vision and Learning Laboratory (BIVL(2)ab), Universidad Industrial de Santander (UIS), Colombia.
| |
Collapse
|
3
|
Fernandez-Martín C, Silva-Rodriguez J, Kiraz U, Morales S, Janssen EAM, Naranjo V. Uninformed Teacher-Student for hard-samples distillation in weakly supervised mitosis localization. Comput Med Imaging Graph 2024; 112:102328. [PMID: 38244279 DOI: 10.1016/j.compmedimag.2024.102328] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/12/2023] [Revised: 11/02/2023] [Accepted: 12/12/2023] [Indexed: 01/22/2024]
Abstract
BACKGROUND AND OBJECTIVE Mitotic activity is a crucial biomarker for diagnosing and predicting outcomes for different types of cancers, particularly breast cancer. However, manual mitosis counting is challenging and time-consuming for pathologists, with moderate reproducibility due to biopsy slide size, low mitotic cell density, and pattern heterogeneity. In recent years, deep learning methods based on convolutional neural networks (CNNs) have been proposed to address these limitations. Nonetheless, these methods have been hampered by the available data labels, which usually consist only of the centroids of mitosis, and by the incoming noise from annotated hard negatives. As a result, complex algorithms with multiple stages are often required to refine the labels at the pixel level and reduce the number of false positives. METHODS This article presents a novel weakly supervised approach for mitosis detection that utilizes only image-level labels on histological hematoxylin and eosin (H&E) images, avoiding the need for complex labeling scenarios. Also, an Uninformed Teacher-Student (UTS) pipeline is introduced to detect and distill hard samples by comparing weakly supervised localizations and the annotated centroids, using strong augmentations to enhance uncertainty. Additionally, an automatic proliferation score is proposed that mimicks the pathologist-annotated mitotic activity index (MAI). The proposed approach is evaluated on three publicly available datasets for mitosis detection on breast histology samples, and two datasets for mitotic activity counting in whole-slide images. RESULTS The proposed framework achieves competitive performance with relevant prior literature in all the datasets used for evaluation without explicitly using the mitosis location information during training. This approach challenges previous methods that rely on strong mitosis location information and multiple stages to refine false positives. Furthermore, the proposed pipeline for hard-sample distillation demonstrates promising dataset-specific improvements. Concretely, when the annotation has not been thoroughly refined by multiple pathologists, the UTS model offers improvements of up to ∼4% in mitosis localization, thanks to the detection and distillation of uncertain cases. Concerning the mitosis counting task, the proposed automatic proliferation score shows a moderate positive correlation with the MAI annotated by pathologists at the biopsy level on two external datasets. CONCLUSIONS The proposed Uninformed Teacher-Student pipeline leverages strong augmentations to distill uncertain samples and measure dissimilarities between predicted and annotated mitosis. Results demonstrate the feasibility of the weakly supervised approach and highlight its potential as an objective evaluation tool for tumor proliferation.
Collapse
Affiliation(s)
- Claudio Fernandez-Martín
- Instituto Universitario de Investigación en Tecnología Centrada en el Ser Humano, HUMAN-tech, Universitat Politècnica de València, Valencia, Spain.
| | | | - Umay Kiraz
- Department of Chemistry, Bioscience and Environmental Engineering, University of Stavanger, Stavanger, Norway; Department of Pathology, Stavanger University Hospital, Stavanger, Norway
| | - Sandra Morales
- Instituto Universitario de Investigación en Tecnología Centrada en el Ser Humano, HUMAN-tech, Universitat Politècnica de València, Valencia, Spain
| | - Emiel A M Janssen
- Department of Chemistry, Bioscience and Environmental Engineering, University of Stavanger, Stavanger, Norway; Department of Pathology, Stavanger University Hospital, Stavanger, Norway
| | - Valery Naranjo
- Instituto Universitario de Investigación en Tecnología Centrada en el Ser Humano, HUMAN-tech, Universitat Politècnica de València, Valencia, Spain
| |
Collapse
|
4
|
Shi Y, Wang H, Ji H, Liu H, Li Y, He N, Wei D, Huang Y, Dai Q, Wu J, Chen X, Zheng Y, Yu H. A deep weakly semi-supervised framework for endoscopic lesion segmentation. Med Image Anal 2023; 90:102973. [PMID: 37757643 DOI: 10.1016/j.media.2023.102973] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 09/05/2022] [Revised: 07/19/2023] [Accepted: 09/14/2023] [Indexed: 09/29/2023]
Abstract
In the field of medical image analysis, accurate lesion segmentation is beneficial for the subsequent clinical diagnosis and treatment planning. Currently, various deep learning-based methods have been proposed to deal with the segmentation task. Albeit achieving some promising performances, the fully-supervised learning approaches require pixel-level annotations for model training, which is tedious and time-consuming for experienced radiologists to collect. In this paper, we propose a weakly semi-supervised segmentation framework, called Point Segmentation Transformer (Point SEGTR). Particularly, the framework utilizes a small amount of fully-supervised data with pixel-level segmentation masks and a large amount of weakly-supervised data with point-level annotations (i.e., annotating a point inside each object) for network training, which largely reduces the demand of pixel-level annotations significantly. To fully exploit the pixel-level and point-level annotations, we propose two regularization terms, i.e., multi-point consistency and symmetric consistency, to boost the quality of pseudo labels, which are then adopted to train a student model for inference. Extensive experiments are conducted on three endoscopy datasets with different lesion structures and several body sites (e.g., colorectal and nasopharynx). Comprehensive experimental results finely substantiate the effectiveness and the generality of our proposed method, as well as its potential to loosen the requirements of pixel-level annotations, which is valuable for clinical applications.
Collapse
Affiliation(s)
- Yuxuan Shi
- ENT Institute and Department of Otolaryngology, Eye & ENT Hospital, Fudan University, Shanghai, 200031, China
| | - Hong Wang
- Tencent Jarvis Lab, Shenzhen 518000, China.
| | - Haoqin Ji
- Tencent Jarvis Lab, Shenzhen 518000, China
| | - Haozhe Liu
- Tencent Jarvis Lab, Shenzhen 518000, China
| | | | - Nanjun He
- Tencent Jarvis Lab, Shenzhen 518000, China
| | - Dong Wei
- Tencent Jarvis Lab, Shenzhen 518000, China
| | | | - Qi Dai
- ENT Institute and Department of Otolaryngology, Eye & ENT Hospital, Fudan University, Shanghai, 200031, China
| | - Jianrong Wu
- Tencent Healthcare (Shenzhen) Co. LTD., Shenzhen 518063, China
| | - Xinrong Chen
- Academy for Engineering and Technology, Fudan University, 220 Handan Road, Shanghai 200033, China.
| | | | - Hongmeng Yu
- ENT Institute and Department of Otolaryngology, Eye & ENT Hospital, Fudan University, Shanghai, 200031, China; Research Units of New Technologies of Endoscopic Surgery in Skull Base Tumor, Chinese Academy of Medical Sciences, 2018RU003, China.
| |
Collapse
|
5
|
Lindemann MC, Glänzer L, Roeth AA, Schmitz-Rode T, Slabu I. Towards Realistic 3D Models of Tumor Vascular Networks. Cancers (Basel) 2023; 15:5352. [PMID: 38001612 PMCID: PMC10670125 DOI: 10.3390/cancers15225352] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 08/28/2023] [Revised: 11/03/2023] [Accepted: 11/03/2023] [Indexed: 11/26/2023] Open
Abstract
For reliable in silico or in vitro investigations in, for example, biosensing and drug delivery applications, accurate models of tumor vascular networks down to the capillary size are essential. Compared to images acquired with conventional medical imaging techniques, digitalized histological tumor slices have a higher resolution, enabling the delineation of capillaries. Volume rendering procedures can then be used to generate a 3D model. However, the preparation of such slices leads to misalignments in relative slice orientation between consecutive slices. Thus, image registration algorithms are necessary to re-align the slices. Here, we present an algorithm for the registration and reconstruction of a vascular network from histologic slices applied to 169 tumor slices. The registration includes two steps. First, consecutive images are incrementally pre-aligned using feature- and area-based transformations. Second, using the previous transformations, parallel registration for all images is enabled. Combining intensity- and color-based thresholds along with heuristic analysis, vascular structures are segmented. A 3D interpolation technique is used for volume rendering. This results in a 3D vascular network with approximately 400-450 vessels with diameters down to 25-30 µm. A delineation of vessel structures with close distance was limited in areas of high structural density. Improvement can be achieved by using images with higher resolution and or machine learning techniques.
Collapse
Affiliation(s)
- Max C. Lindemann
- Institute of Applied Medical Engineering, Helmholtz Institute, Medical Faculty, RWTH Aachen University, Pauwelsstraße 20, 52074 Aachen, Germany (L.G.); (T.S.-R.)
| | - Lukas Glänzer
- Institute of Applied Medical Engineering, Helmholtz Institute, Medical Faculty, RWTH Aachen University, Pauwelsstraße 20, 52074 Aachen, Germany (L.G.); (T.S.-R.)
| | - Anjali A. Roeth
- Department of General, Visceral and Transplant Surgery, RWTH Aachen University Hospital, Pauwelsstrasse 30, 52074 Aachen, Germany
- Department of Surgery, Maastricht University, P. Debyelaan 25, 6229 HX Maastricht, The Netherlands
| | - Thomas Schmitz-Rode
- Institute of Applied Medical Engineering, Helmholtz Institute, Medical Faculty, RWTH Aachen University, Pauwelsstraße 20, 52074 Aachen, Germany (L.G.); (T.S.-R.)
| | - Ioana Slabu
- Institute of Applied Medical Engineering, Helmholtz Institute, Medical Faculty, RWTH Aachen University, Pauwelsstraße 20, 52074 Aachen, Germany (L.G.); (T.S.-R.)
| |
Collapse
|
6
|
Pati P, Jaume G, Ayadi Z, Thandiackal K, Bozorgtabar B, Gabrani M, Goksel O. Weakly supervised joint whole-slide segmentation and classification in prostate cancer. Med Image Anal 2023; 89:102915. [PMID: 37633177 DOI: 10.1016/j.media.2023.102915] [Citation(s) in RCA: 7] [Impact Index Per Article: 7.0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 09/26/2022] [Revised: 05/17/2023] [Accepted: 07/25/2023] [Indexed: 08/28/2023]
Abstract
The identification and segmentation of histological regions of interest can provide significant support to pathologists in their diagnostic tasks. However, segmentation methods are constrained by the difficulty in obtaining pixel-level annotations, which are tedious and expensive to collect for whole-slide images (WSI). Though several methods have been developed to exploit image-level weak-supervision for WSI classification, the task of segmentation using WSI-level labels has received very little attention. The research in this direction typically require additional supervision beyond image labels, which are difficult to obtain in real-world practice. In this study, we propose WholeSIGHT, a weakly-supervised method that can simultaneously segment and classify WSIs of arbitrary shapes and sizes. Formally, WholeSIGHT first constructs a tissue-graph representation of WSI, where the nodes and edges depict tissue regions and their interactions, respectively. During training, a graph classification head classifies the WSI and produces node-level pseudo-labels via post-hoc feature attribution. These pseudo-labels are then used to train a node classification head for WSI segmentation. During testing, both heads simultaneously render segmentation and class prediction for an input WSI. We evaluate the performance of WholeSIGHT on three public prostate cancer WSI datasets. Our method achieves state-of-the-art weakly-supervised segmentation performance on all datasets while resulting in better or comparable classification with respect to state-of-the-art weakly-supervised WSI classification methods. Additionally, we assess the generalization capability of our method in terms of segmentation and classification performance, uncertainty estimation, and model calibration. Our code is available at: https://github.com/histocartography/wholesight.
Collapse
Affiliation(s)
| | - Guillaume Jaume
- Department of Pathology, Brigham and Women's Hospital, Harvard Medical School, Boston, MA, USA; Department of Pathology, Massachusetts General Hospital, Harvard Medical School, Boston, MA, USA; Cancer Program, Broad Institute of Harvard and MIT, Cambridge, MA, USA; Data Science Program, Dana-Farber/Harvard Cancer Center, Boston, MA, USA
| | - Zeineb Ayadi
- IBM Research Europe, Zurich, Switzerland; EPFL, Lausanne, Switzerland
| | - Kevin Thandiackal
- IBM Research Europe, Zurich, Switzerland; Computer-Assisted Applications in Medicine, ETH Zurich, Zurich, Switzerland
| | | | | | - Orcun Goksel
- Computer-Assisted Applications in Medicine, ETH Zurich, Zurich, Switzerland; Department of Information Technology, Uppsala University, Sweden
| |
Collapse
|
7
|
Zheng Y, Huang D, Hao X, Wei J, Lu H, Liu Y. UniVisNet: A Unified Visualization and Classification Network for accurate grading of gliomas from MRI. Comput Biol Med 2023; 165:107332. [PMID: 37598632 DOI: 10.1016/j.compbiomed.2023.107332] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/30/2023] [Revised: 07/30/2023] [Accepted: 08/07/2023] [Indexed: 08/22/2023]
Abstract
Accurate grading of brain tumors plays a crucial role in the diagnosis and treatment of glioma. While convolutional neural networks (CNNs) have shown promising performance in this task, their clinical applicability is still constrained by the interpretability and robustness of the models. In the conventional framework, the classification model is trained first, and then visual explanations are generated. However, this approach often leads to models that prioritize classification performance or complexity, making it difficult to achieve a precise visual explanation. Motivated by these challenges, we propose the Unified Visualization and Classification Network (UniVisNet), a novel framework that aims to improve both the classification performance and the generation of high-resolution visual explanations. UniVisNet addresses attention misalignment by introducing a subregion-based attention mechanism, which replaces traditional down-sampling operations. Additionally, multiscale feature maps are fused to achieve higher resolution, enabling the generation of detailed visual explanations. To streamline the process, we introduce the Unified Visualization and Classification head (UniVisHead), which directly generates visual explanations without the need for additional separation steps. Through extensive experiments, our proposed UniVisNet consistently outperforms strong baseline classification models and prevalent visualization methods. Notably, UniVisNet achieves remarkable results on the glioma grading task, including an AUC of 94.7%, an accuracy of 89.3%, a sensitivity of 90.4%, and a specificity of 85.3%. Moreover, UniVisNet provides visually interpretable explanations that surpass existing approaches. In conclusion, UniVisNet innovatively generates visual explanations in brain tumor grading by simultaneously improving the classification performance and generating high-resolution visual explanations. This work contributes to the clinical application of deep learning, empowering clinicians with comprehensive insights into the spatial heterogeneity of glioma.
Collapse
Affiliation(s)
- Yao Zheng
- Air Force Medical University, No. 169 Changle West Road, Xi'an, 710032, ShaanXi, China
| | - Dong Huang
- Air Force Medical University, No. 169 Changle West Road, Xi'an, 710032, ShaanXi, China; Shaanxi Provincial Key Laboratory of Bioelectromagnetic Detection and Intelligent Perception, No. 169 Changle West Road, Xi'an, 710032, ShaanXi, China
| | - Xiaoshuo Hao
- Air Force Medical University, No. 169 Changle West Road, Xi'an, 710032, ShaanXi, China
| | - Jie Wei
- Air Force Medical University, No. 169 Changle West Road, Xi'an, 710032, ShaanXi, China
| | - Hongbing Lu
- Air Force Medical University, No. 169 Changle West Road, Xi'an, 710032, ShaanXi, China; Shaanxi Provincial Key Laboratory of Bioelectromagnetic Detection and Intelligent Perception, No. 169 Changle West Road, Xi'an, 710032, ShaanXi, China.
| | - Yang Liu
- Air Force Medical University, No. 169 Changle West Road, Xi'an, 710032, ShaanXi, China; Shaanxi Provincial Key Laboratory of Bioelectromagnetic Detection and Intelligent Perception, No. 169 Changle West Road, Xi'an, 710032, ShaanXi, China.
| |
Collapse
|
8
|
Jing Y, Li C, Du T, Jiang T, Sun H, Yang J, Shi L, Gao M, Grzegorzek M, Li X. A comprehensive survey of intestine histopathological image analysis using machine vision approaches. Comput Biol Med 2023; 165:107388. [PMID: 37696178 DOI: 10.1016/j.compbiomed.2023.107388] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 06/08/2023] [Revised: 08/06/2023] [Accepted: 08/25/2023] [Indexed: 09/13/2023]
Abstract
Colorectal Cancer (CRC) is currently one of the most common and deadly cancers. CRC is the third most common malignancy and the fourth leading cause of cancer death worldwide. It ranks as the second most frequent cause of cancer-related deaths in the United States and other developed countries. Histopathological images contain sufficient phenotypic information, they play an indispensable role in the diagnosis and treatment of CRC. In order to improve the objectivity and diagnostic efficiency for image analysis of intestinal histopathology, Computer-aided Diagnosis (CAD) methods based on machine learning (ML) are widely applied in image analysis of intestinal histopathology. In this investigation, we conduct a comprehensive study on recent ML-based methods for image analysis of intestinal histopathology. First, we discuss commonly used datasets from basic research studies with knowledge of intestinal histopathology relevant to medicine. Second, we introduce traditional ML methods commonly used in intestinal histopathology, as well as deep learning (DL) methods. Then, we provide a comprehensive review of the recent developments in ML methods for segmentation, classification, detection, and recognition, among others, for histopathological images of the intestine. Finally, the existing methods have been studied, and the application prospects of these methods in this field are given.
Collapse
Affiliation(s)
- Yujie Jing
- Microscopic Image and Medical Image Analysis Group, College of Medicine and Biological Information Engineering, Northeastern University, China; Key Laboratory of Intelligent Computing in Medical Image, Ministry of Education, Northeastern University, Shenyang, Liaoning, China
| | - Chen Li
- Microscopic Image and Medical Image Analysis Group, College of Medicine and Biological Information Engineering, Northeastern University, China; Key Laboratory of Intelligent Computing in Medical Image, Ministry of Education, Northeastern University, Shenyang, Liaoning, China.
| | - Tianming Du
- Microscopic Image and Medical Image Analysis Group, College of Medicine and Biological Information Engineering, Northeastern University, China; Key Laboratory of Intelligent Computing in Medical Image, Ministry of Education, Northeastern University, Shenyang, Liaoning, China
| | - Tao Jiang
- School of Intelligent Medicine, Chengdu University of Traditional Chinese Medicine, Chengdu, China; International Joint Institute of Robotics and Intelligent Systems, Chengdu University of Information Technology, Chengdu, China
| | - Hongzan Sun
- Shengjing Hospital of China Medical University, Shenyang, China
| | - Jinzhu Yang
- Key Laboratory of Intelligent Computing in Medical Image, Ministry of Education, Northeastern University, Shenyang, Liaoning, China
| | - Liyu Shi
- Microscopic Image and Medical Image Analysis Group, College of Medicine and Biological Information Engineering, Northeastern University, China; Key Laboratory of Intelligent Computing in Medical Image, Ministry of Education, Northeastern University, Shenyang, Liaoning, China
| | - Minghe Gao
- Microscopic Image and Medical Image Analysis Group, College of Medicine and Biological Information Engineering, Northeastern University, China; Key Laboratory of Intelligent Computing in Medical Image, Ministry of Education, Northeastern University, Shenyang, Liaoning, China
| | - Marcin Grzegorzek
- Institute for Medical Informatics, University of Luebeck, Luebeck, Germany; Department of Knowledge Engineering, University of Economics in Katowice, Katowice, Poland
| | - Xiaoyan Li
- Cancer Hospital of China Medical University, Liaoning Cancer Hospital, Shenyang, China.
| |
Collapse
|
9
|
Zheng T, Chen W, Li S, Quan H, Zou M, Zheng S, Zhao Y, Gao X, Cui X. Learning how to detect: A deep reinforcement learning method for whole-slide melanoma histopathology images. Comput Med Imaging Graph 2023; 108:102275. [PMID: 37567046 DOI: 10.1016/j.compmedimag.2023.102275] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 04/08/2023] [Revised: 07/18/2023] [Accepted: 07/22/2023] [Indexed: 08/13/2023]
Abstract
Cutaneous melanoma represents one of the most life-threatening malignancies. Histopathological image analysis serves as a vital tool for early melanoma detection. Deep neural network (DNN) models are frequently employed to aid pathologists in enhancing the efficiency and accuracy of diagnoses. However, due to the paucity of well-annotated, high-resolution, whole-slide histopathology image (WSI) datasets, WSIs are typically fragmented into numerous patches during the model training and testing stages. This process disregards the inherent interconnectedness among patches, potentially impeding the models' performance. Additionally, the presence of excess, non-contributing patches extends processing times and introduces substantial computational burdens. To mitigate these issues, we draw inspiration from the clinical decision-making processes of dermatopathologists to propose an innovative, weakly supervised deep reinforcement learning framework, titled Fast medical decision-making in melanoma histopathology images (FastMDP-RL). This framework expedites model inference by reducing the number of irrelevant patches identified within WSIs. FastMDP-RL integrates two DNN-based agents: the search agent (SeAgent) and the decision agent (DeAgent). The SeAgent initiates actions, steered by the image features observed in the current viewing field at various magnifications. Simultaneously, the DeAgent provides labeling probabilities for each patch. We utilize multi-instance learning (MIL) to construct a teacher-guided model (MILTG), serving a dual purpose: rewarding the SeAgent and guiding the DeAgent. Our evaluations were conducted using two melanoma datasets: the publicly accessible TCIA-CM dataset and the proprietary MELSC dataset. Our experimental findings affirm FastMDP-RL's ability to expedite inference and accurately predict WSIs, even in the absence of pixel-level annotations. Moreover, our research investigates the WSI-based interactive environment, encompassing the design of agents, state and reward functions, and feature extractors suitable for melanoma tissue images. This investigation offers valuable insights and references for researchers engaged in related studies. The code is available at: https://github.com/titizheng/FastMDP-RL.
Collapse
Affiliation(s)
- Tingting Zheng
- College of Medicine and Biological Information Engineering, Northeastern University, Shenyang, China
| | - Weixing Chen
- Shenzhen College of Advanced Technology, University of the Chinese Academy of Sciences, Beijing, China
| | - Shuqin Li
- College of Medicine and Biological Information Engineering, Northeastern University, Shenyang, China
| | - Hao Quan
- College of Medicine and Biological Information Engineering, Northeastern University, Shenyang, China
| | - Mingchen Zou
- College of Medicine and Biological Information Engineering, Northeastern University, Shenyang, China
| | - Song Zheng
- National and Local Joint Engineering Research Center of Immunodermatological Theranostics, Department of Dermatology, The First Hospital of China Medical University, Shenyang, China
| | - Yue Zhao
- College of Medicine and Biological Information Engineering, Northeastern University, Shenyang, China; National and Local Joint Engineering Research Center of Immunodermatological Theranostics, Department of Dermatology, The First Hospital of China Medical University, Shenyang, China
| | - Xinghua Gao
- National and Local Joint Engineering Research Center of Immunodermatological Theranostics, Department of Dermatology, The First Hospital of China Medical University, Shenyang, China
| | - Xiaoyu Cui
- College of Medicine and Biological Information Engineering, Northeastern University, Shenyang, China.
| |
Collapse
|
10
|
Ramesh S, Dall'Alba D, Gonzalez C, Yu T, Mascagni P, Mutter D, Marescaux J, Fiorini P, Padoy N. Weakly Supervised Temporal Convolutional Networks for Fine-Grained Surgical Activity Recognition. IEEE TRANSACTIONS ON MEDICAL IMAGING 2023; 42:2592-2602. [PMID: 37030859 DOI: 10.1109/tmi.2023.3262847] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 06/19/2023]
Abstract
Automatic recognition of fine-grained surgical activities, called steps, is a challenging but crucial task for intelligent intra-operative computer assistance. The development of current vision-based activity recognition methods relies heavily on a high volume of manually annotated data. This data is difficult and time-consuming to generate and requires domain-specific knowledge. In this work, we propose to use coarser and easier-to-annotate activity labels, namely phases, as weak supervision to learn step recognition with fewer step annotated videos. We introduce a step-phase dependency loss to exploit the weak supervision signal. We then employ a Single-Stage Temporal Convolutional Network (SS-TCN) with a ResNet-50 backbone, trained in an end-to-end fashion from weakly annotated videos, for temporal activity segmentation and recognition. We extensively evaluate and show the effectiveness of the proposed method on a large video dataset consisting of 40 laparoscopic gastric bypass procedures and the public benchmark CATARACTS containing 50 cataract surgeries.
Collapse
|
11
|
Liu Z, Lv Q, Yang Z, Li Y, Lee CH, Shen L. Recent progress in transformer-based medical image analysis. Comput Biol Med 2023; 164:107268. [PMID: 37494821 DOI: 10.1016/j.compbiomed.2023.107268] [Citation(s) in RCA: 10] [Impact Index Per Article: 10.0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/21/2023] [Revised: 05/30/2023] [Accepted: 07/16/2023] [Indexed: 07/28/2023]
Abstract
The transformer is primarily used in the field of natural language processing. Recently, it has been adopted and shows promise in the computer vision (CV) field. Medical image analysis (MIA), as a critical branch of CV, also greatly benefits from this state-of-the-art technique. In this review, we first recap the core component of the transformer, the attention mechanism, and the detailed structures of the transformer. After that, we depict the recent progress of the transformer in the field of MIA. We organize the applications in a sequence of different tasks, including classification, segmentation, captioning, registration, detection, enhancement, localization, and synthesis. The mainstream classification and segmentation tasks are further divided into eleven medical image modalities. A large number of experiments studied in this review illustrate that the transformer-based method outperforms existing methods through comparisons with multiple evaluation metrics. Finally, we discuss the open challenges and future opportunities in this field. This task-modality review with the latest contents, detailed information, and comprehensive comparison may greatly benefit the broad MIA community.
Collapse
Affiliation(s)
- Zhaoshan Liu
- Department of Mechanical Engineering, National University of Singapore, 9 Engineering Drive 1, Singapore, 117575, Singapore.
| | - Qiujie Lv
- Department of Mechanical Engineering, National University of Singapore, 9 Engineering Drive 1, Singapore, 117575, Singapore; School of Intelligent Systems Engineering, Sun Yat-sen University, No. 66, Gongchang Road, Guangming District, 518107, China.
| | - Ziduo Yang
- Department of Mechanical Engineering, National University of Singapore, 9 Engineering Drive 1, Singapore, 117575, Singapore; School of Intelligent Systems Engineering, Sun Yat-sen University, No. 66, Gongchang Road, Guangming District, 518107, China.
| | - Yifan Li
- Department of Mechanical Engineering, National University of Singapore, 9 Engineering Drive 1, Singapore, 117575, Singapore.
| | - Chau Hung Lee
- Department of Radiology, Tan Tock Seng Hospital, 11 Jalan Tan Tock Seng, Singapore, 308433, Singapore.
| | - Lei Shen
- Department of Mechanical Engineering, National University of Singapore, 9 Engineering Drive 1, Singapore, 117575, Singapore.
| |
Collapse
|
12
|
Khazaee Fadafen M, Rezaee K. Ensemble-based multi-tissue classification approach of colorectal cancer histology images using a novel hybrid deep learning framework. Sci Rep 2023; 13:8823. [PMID: 37258631 DOI: 10.1038/s41598-023-35431-x] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 04/04/2022] [Accepted: 05/17/2023] [Indexed: 06/02/2023] Open
Abstract
Colorectal cancer (CRC) is the second leading cause of cancer death in the world, so digital pathology is essential for assessing prognosis. Due to the increasing resolution and quantity of whole slide images (WSIs), as well as the lack of annotated information, previous methodologies cannot be generalized as effective decision-making systems. Since deep learning (DL) methods can handle large-scale applications, they can provide a viable alternative to histopathology image (HI) analysis. DL architectures, however, may not be sufficient to classify CRC tissues based on anatomical histopathology data. A dilated ResNet (dResNet) structure and attention module are used to generate deep feature maps in order to classify multiple tissues in HIs. In addition, neighborhood component analysis (NCA) overcomes the constraint of computational complexity. Data is fed into a deep support vector machine (SVM) based on an ensemble learning algorithm called DeepSVM after the features have been selected. CRC-5000 and NCT-CRC-HE-100 K datasets were analyzed to validate and test the hybrid procedure. We demonstrate that the hybrid model achieves 98.75% and 99.76% accuracy on CRC datasets. The results showed that only pathologists' labels could successfully classify unseen WSIs. Furthermore, the hybrid deep learning method outperforms state-of-the-art approaches in terms of computational efficiency and time. Using the proposed mechanism for tissue analysis, it will be possible to correctly predict CRC based on accurate pathology image classification.
Collapse
Affiliation(s)
- Masoud Khazaee Fadafen
- Department of Electrical Engineering, Technical and Vocational University (TVU), Tehran, Iran
| | - Khosro Rezaee
- Department of Biomedical Engineering, Meybod University, Meybod, Iran.
| |
Collapse
|
13
|
Hu H, Ye R, Thiyagalingam J, Coenen F, Su J. Triple-kernel gated attention-based multiple instance learning with contrastive learning for medical image analysis. APPL INTELL 2023; 53:1-16. [PMID: 37363384 PMCID: PMC10072016 DOI: 10.1007/s10489-023-04458-y] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Accepted: 01/05/2023] [Indexed: 04/07/2023]
Abstract
In machine learning, multiple instance learning is a method evolved from supervised learning algorithms, which defines a "bag" as a collection of multiple examples with a wide range of applications. In this paper, we propose a novel deep multiple instance learning model for medical image analysis, called triple-kernel gated attention-based multiple instance learning with contrastive learning. It can be used to overcome the limitations of the existing multiple instance learning approaches to medical image analysis. Our model consists of four steps. i) Extracting the representations by a simple convolutional neural network using contrastive learning for training. ii) Using three different kernel functions to obtain the importance of each instance from the entire image and forming an attention map. iii) Based on the attention map, aggregating the entire image together by attention-based MIL pooling. iv) Feeding the results into the classifier for prediction. The results on different datasets demonstrate that the proposed model outperforms state-of-the-art methods on binary and weakly supervised classification tasks. It can provide more efficient classification results for various disease models and additional explanatory information.
Collapse
Affiliation(s)
- Huafeng Hu
- Department of Electrical and Electronic Engineering, University of Liverpool based at Xi’an Jiaotong-Liverpool University, Suzhou, 215123 Jiangsu China
| | - Ruijie Ye
- Department of Computer Science, University of Liverpool, Liverpool, L69 3BX UK
| | - Jeyan Thiyagalingam
- Scientific Computing Department, Rutherford Appleton Laboratory, Science and Technology Facilities Council, Harwell Campus, Didcot, OX11 0QX UK
| | - Frans Coenen
- Department of Computer Science, University of Liverpool, Liverpool, L69 3BX UK
| | - Jionglong Su
- School of AI and Advanced Computing, XJTLU Entrepreneur College (Taicang), Xi’an Jiaotong-Liverpool University, Suzhou, 215123 Jiangsu China
| |
Collapse
|
14
|
Li K, Qian Z, Han Y, Chang EIC, Wei B, Lai M, Liao J, Fan Y, Xu Y. Weakly supervised histopathology image segmentation with self-attention. Med Image Anal 2023; 86:102791. [PMID: 36933385 DOI: 10.1016/j.media.2023.102791] [Citation(s) in RCA: 2] [Impact Index Per Article: 2.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 01/21/2022] [Revised: 01/09/2023] [Accepted: 02/24/2023] [Indexed: 03/13/2023]
Abstract
Accurate segmentation in histopathology images at pixel-level plays a critical role in the digital pathology workflow. The development of weakly supervised methods for histopathology image segmentation liberates pathologists from time-consuming and labor-intensive works, opening up possibilities of further automated quantitative analysis of whole-slide histopathology images. As an effective subgroup of weakly supervised methods, multiple instance learning (MIL) has achieved great success in histopathology images. In this paper, we specially treat pixels as instances so that the histopathology image segmentation task is transformed into an instance prediction task in MIL. However, the lack of relations between instances in MIL limits the further improvement of segmentation performance. Therefore, we propose a novel weakly supervised method called SA-MIL for pixel-level segmentation in histopathology images. SA-MIL introduces a self-attention mechanism into the MIL framework, which captures global correlation among all instances. In addition, we use deep supervision to make the best use of information from limited annotations in the weakly supervised method. Our approach makes up for the shortcoming that instances are independent of each other in MIL by aggregating global contextual information. We demonstrate state-of-the-art results compared to other weakly supervised methods on two histopathology image datasets. It is evident that our approach has generalization ability for the high performance on both tissue and cell histopathology datasets. There is potential in our approach for various applications in medical images.
Collapse
Affiliation(s)
- Kailu Li
- School of Biological Science and Medical Engineering, State Key Laboratory of Software Development Environment, Key Laboratory of Biomechanics, Mechanobiology of Ministry of Education and Beijing Advanced Innovation Centre for Biomedical Engineering, Beihang University, Beijing 100191, China.
| | - Ziniu Qian
- School of Biological Science and Medical Engineering, State Key Laboratory of Software Development Environment, Key Laboratory of Biomechanics, Mechanobiology of Ministry of Education and Beijing Advanced Innovation Centre for Biomedical Engineering, Beihang University, Beijing 100191, China.
| | - Yingnan Han
- School of Biological Science and Medical Engineering, State Key Laboratory of Software Development Environment, Key Laboratory of Biomechanics, Mechanobiology of Ministry of Education and Beijing Advanced Innovation Centre for Biomedical Engineering, Beihang University, Beijing 100191, China.
| | | | | | - Maode Lai
- Department of Pathology, School of Medicine, Zhejiang University, Hangzhou 310027, China.
| | - Jing Liao
- Department of Computer Science, City University of Hong Kong, 999077, Hong Kong SAR, China.
| | - Yubo Fan
- School of Biological Science and Medical Engineering, State Key Laboratory of Software Development Environment, Key Laboratory of Biomechanics, Mechanobiology of Ministry of Education and Beijing Advanced Innovation Centre for Biomedical Engineering, Beihang University, Beijing 100191, China.
| | - Yan Xu
- School of Biological Science and Medical Engineering, State Key Laboratory of Software Development Environment, Key Laboratory of Biomechanics, Mechanobiology of Ministry of Education and Beijing Advanced Innovation Centre for Biomedical Engineering, Beihang University, Beijing 100191, China; Microsoft Research, Beijing 100080, China.
| |
Collapse
|
15
|
Manescu P, Narayanan P, Bendkowski C, Elmi M, Claveau R, Pawar V, Brown BJ, Shaw M, Rao A, Fernandez-Reyes D. Detection of acute promyelocytic leukemia in peripheral blood and bone marrow with annotation-free deep learning. Sci Rep 2023; 13:2562. [PMID: 36781917 PMCID: PMC9925435 DOI: 10.1038/s41598-023-29160-4] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 08/30/2022] [Accepted: 01/31/2023] [Indexed: 02/15/2023] Open
Abstract
While optical microscopy inspection of blood films and bone marrow aspirates by a hematologist is a crucial step in establishing diagnosis of acute leukemia, especially in low-resource settings where other diagnostic modalities are not available, the task remains time-consuming and prone to human inconsistencies. This has an impact especially in cases of Acute Promyelocytic Leukemia (APL) that require urgent treatment. Integration of automated computational hematopathology into clinical workflows can improve the throughput of these services and reduce cognitive human error. However, a major bottleneck in deploying such systems is a lack of sufficient cell morphological object-labels annotations to train deep learning models. We overcome this by leveraging patient diagnostic labels to train weakly-supervised models that detect different types of acute leukemia. We introduce a deep learning approach, Multiple Instance Learning for Leukocyte Identification (MILLIE), able to perform automated reliable analysis of blood films with minimal supervision. Without being trained to classify individual cells, MILLIE differentiates between acute lymphoblastic and myeloblastic leukemia in blood films. More importantly, MILLIE detects APL in blood films (AUC 0.94 ± 0.04) and in bone marrow aspirates (AUC 0.99 ± 0.01). MILLIE is a viable solution to augment the throughput of clinical pathways that require assessment of blood film microscopy.
Collapse
Affiliation(s)
- Petru Manescu
- Department of Computer Science, Faculty of Engineering Sciences, University College London, Gower Street, London, WC1E 6BT, UK.
| | - Priya Narayanan
- Department of Computer Science, Faculty of Engineering Sciences, University College London, Gower Street, London, WC1E 6BT, UK
| | - Christopher Bendkowski
- Department of Computer Science, Faculty of Engineering Sciences, University College London, Gower Street, London, WC1E 6BT, UK
| | - Muna Elmi
- Department of Computer Science, Faculty of Engineering Sciences, University College London, Gower Street, London, WC1E 6BT, UK
| | - Remy Claveau
- Department of Computer Science, Faculty of Engineering Sciences, University College London, Gower Street, London, WC1E 6BT, UK
| | - Vijay Pawar
- Department of Computer Science, Faculty of Engineering Sciences, University College London, Gower Street, London, WC1E 6BT, UK
| | - Biobele J Brown
- Department of Paediatrics, College of Medicine, University of Ibadan, University College Hospital, Ibadan, Nigeria
| | - Mike Shaw
- Department of Computer Science, Faculty of Engineering Sciences, University College London, Gower Street, London, WC1E 6BT, UK
| | - Anupama Rao
- Department of Haematology, Great Ormond Street Hospital for Children, London, WC1N 3JH, UK
| | - Delmiro Fernandez-Reyes
- Department of Computer Science, Faculty of Engineering Sciences, University College London, Gower Street, London, WC1E 6BT, UK.
- Department of Paediatrics, College of Medicine, University of Ibadan, University College Hospital, Ibadan, Nigeria.
| |
Collapse
|
16
|
Han Y, Cheng L, Huang G, Zhong G, Li J, Yuan X, Liu H, Li J, Zhou J, Cai M. Weakly supervised semantic segmentation of histological tissue via attention accumulation and pixel-level contrast learning. Phys Med Biol 2023; 68. [PMID: 36577142 DOI: 10.1088/1361-6560/acaeee] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 08/10/2022] [Accepted: 12/28/2022] [Indexed: 12/29/2022]
Abstract
Objective. Histopathology image segmentation can assist medical professionals in identifying and diagnosing diseased tissue more efficiently. Although fully supervised segmentation models have excellent performance, the annotation cost is extremely expensive. Weakly supervised models are widely used in medical image segmentation due to their low annotation cost. Nevertheless, these weakly supervised models have difficulty in accurately locating the boundaries between different classes of regions in pathological images, resulting in a high rate of false alarms Our objective is to design a weakly supervised segmentation model to resolve the above problems.Approach. The segmentation model is divided into two main stages, the generation of pseudo labels based on class residual attention accumulation network (CRAANet) and the semantic segmentation based on pixel feature space construction network (PFSCNet). CRAANet provides attention scores for each class through the class residual attention module, while the Attention Accumulation (AA) module overlays the attention feature maps generated in each training epoch. PFSCNet employs a network model containing an inflated convolutional residual neural network and a multi-scale feature-aware module as the segmentation backbone, and proposes dense energy loss and pixel clustering modules are based on contrast learning to solve the pseudo-labeling-inaccuracy problem.Main results. We validate our method using the lung adenocarcinoma (LUAD-HistoSeg) dataset and the breast cancer (BCSS) dataset. The results of the experiments show that our proposed method outperforms other state-of-the-art methods on both datasets in several metrics. This suggests that it is capable of performing well in a wide variety of histopathological image segmentation tasks.Significance. We propose a weakly supervised semantic segmentation network that achieves approximate fully supervised segmentation performance even in the case of incomplete labels. The proposed AA and pixel-level contrast learning also make the edges more accurate and can well assist pathologists in their research.
Collapse
Affiliation(s)
- Yongqi Han
- School of Computer Science and Technology, Guangdong University of Technology, Guangzhou 510006, People's Republic of China
| | - Lianglun Cheng
- School of Computer Science and Technology, Guangdong University of Technology, Guangzhou 510006, People's Republic of China
| | - Guoheng Huang
- School of Computer Science and Technology, Guangdong University of Technology, Guangzhou 510006, People's Republic of China
| | - Guo Zhong
- School of Information Science and Technology, Guangdong University of Foreign Studies, Guangzhou 510420, People's Republic of China
| | - Jiahua Li
- School of Computer Science and Technology, Guangdong University of Technology, Guangzhou 510006, People's Republic of China
| | - Xiaochen Yuan
- Faculty of Applied Sciences, Macao Polytechnic University, Macao 999078, People's Republic of China
| | - Hongrui Liu
- Department of Industrial and Systems Engineering, San Jose State University, CA 95192, United States of America
| | - Jiao Li
- Department of Radiology, Sun Yat-sen University Cancer Center; State Key Laboratory of Oncology in South China, Collaborative Innovation Center for Cancer Medicine, Guangzhou 510060, People's Republic of China
| | - Jian Zhou
- Department of Medical Imaging, Sun Yat-sen University Cancer Center; State Key Laboratory of Oncology in South China; Collaborative Innovation Center for Cancer Medicine, Guangzhou 510060, People's Republic of China
| | - Muyan Cai
- Department of Pathology, Sun Yat-sen University Cancer Center; State Key Laboratory of Oncology in South China; Collaborative Innovation Center for Cancer Medicine, Guangzhou 510060, People's Republic of China
| |
Collapse
|
17
|
Morano J, Hervella ÁS, Rouco J, Novo J, Fernández-Vigo JI, Ortega M. Weakly-supervised detection of AMD-related lesions in color fundus images using explainable deep learning. COMPUTER METHODS AND PROGRAMS IN BIOMEDICINE 2023; 229:107296. [PMID: 36481530 DOI: 10.1016/j.cmpb.2022.107296] [Citation(s) in RCA: 3] [Impact Index Per Article: 3.0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Subscribe] [Scholar Register] [Received: 03/24/2022] [Revised: 11/16/2022] [Accepted: 11/29/2022] [Indexed: 06/17/2023]
Abstract
BACKGROUND AND OBJECTIVES Age-related macular degeneration (AMD) is a degenerative disorder affecting the macula, a key area of the retina for visual acuity. Nowadays, AMD is the most frequent cause of blindness in developed countries. Although some promising treatments have been proposed that effectively slow down its development, their effectiveness significantly diminishes in the advanced stages. This emphasizes the importance of large-scale screening programs for early detection. Nevertheless, implementing such programs for a disease like AMD is usually unfeasible, since the population at risk is large and the diagnosis is challenging. For the characterization of the disease, clinicians have to identify and localize certain retinal lesions. All this motivates the development of automatic diagnostic methods. In this sense, several works have achieved highly positive results for AMD detection using convolutional neural networks (CNNs). However, none of them incorporates explainability mechanisms linking the diagnosis to its related lesions to help clinicians to better understand the decisions of the models. This is specially relevant, since the absence of such mechanisms limits the application of automatic methods in the clinical practice. In that regard, we propose an explainable deep learning approach for the diagnosis of AMD via the joint identification of its associated retinal lesions. METHODS In our proposal, a CNN with a custom architectural setting is trained end-to-end for the joint identification of AMD and its associated retinal lesions. With the proposed setting, the lesion identification is directly derived from independent lesion activation maps; then, the diagnosis is obtained from the identified lesions. The training is performed end-to-end using image-level labels. Thus, lesion-specific activation maps are learned in a weakly-supervised manner. The provided lesion information is of high clinical interest, as it allows clinicians to assess the developmental stage of the disease. Additionally, the proposed approach allows to explain the diagnosis obtained by the models directly from the identified lesions and their corresponding activation maps. The training data necessary for the approach can be obtained without much extra work on the part of clinicians, since the lesion information is habitually present in medical records. This is an important advantage over other methods, including fully-supervised lesion segmentation methods, which require pixel-level labels whose acquisition is arduous. RESULTS The experiments conducted in 4 different datasets demonstrate that the proposed approach is able to identify AMD and its associated lesions with satisfactory performance. Moreover, the evaluation of the lesion activation maps shows that the models trained using the proposed approach are able to identify the pathological areas within the image and, in most cases, to correctly determine to which lesion they correspond. CONCLUSIONS The proposed approach provides meaningful information-lesion identification and lesion activation maps-that conveniently explains and complements the diagnosis, and is of particular interest to clinicians for the diagnostic process. Moreover, the data needed to train the networks using the proposed approach is commonly easy to obtain, what represents an important advantage in fields with particularly scarce data, such as medical imaging.
Collapse
Affiliation(s)
- José Morano
- Centro de Investigación CITIC, Universidade da Coruña, A Coruña, Spain; VARPA Research Group, Instituto de Investigación Biomédica de A Coruńa (INIBIC), Universidade da Coruña, A Coruña, Spain.
| | - Álvaro S Hervella
- Centro de Investigación CITIC, Universidade da Coruña, A Coruña, Spain; VARPA Research Group, Instituto de Investigación Biomédica de A Coruńa (INIBIC), Universidade da Coruña, A Coruña, Spain.
| | - José Rouco
- Centro de Investigación CITIC, Universidade da Coruña, A Coruña, Spain; VARPA Research Group, Instituto de Investigación Biomédica de A Coruńa (INIBIC), Universidade da Coruña, A Coruña, Spain.
| | - Jorge Novo
- Centro de Investigación CITIC, Universidade da Coruña, A Coruña, Spain; VARPA Research Group, Instituto de Investigación Biomédica de A Coruńa (INIBIC), Universidade da Coruña, A Coruña, Spain.
| | - José I Fernández-Vigo
- Department of Ophthalmology, Hospital Clínico San Carlos, Instituto de Investigación Sanitaria (IdISSC), Madrid, Spain; Department of Ophthalmology, Centro Internacional de Oftalmología Avanzada, Madrid, Spain.
| | - Marcos Ortega
- Centro de Investigación CITIC, Universidade da Coruña, A Coruña, Spain; VARPA Research Group, Instituto de Investigación Biomédica de A Coruńa (INIBIC), Universidade da Coruña, A Coruña, Spain.
| |
Collapse
|
18
|
Chen Z, Zheng W, Pang K, Xia D, Guo L, Chen X, Wu F, Wang H. Weakly supervised learning analysis of Aβ plaque distribution in the whole rat brain. Front Neurosci 2023; 16:1097019. [PMID: 36741048 PMCID: PMC9892753 DOI: 10.3389/fnins.2022.1097019] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/13/2022] [Accepted: 12/30/2022] [Indexed: 01/20/2023] Open
Abstract
Alzheimer's disease (AD) is a great challenge for the world and hardly to be cured, partly because of the lack of animal models that fully mimic pathological progress. Recently, a rat model exhibiting the most pathological symptoms of AD has been reported. However, high-resolution imaging and accurate quantification of beta-amyloid (Aβ) plaques in the whole rat brain have not been fulfilled due to substantial technical challenges. In this paper, a high-efficiency data analysis pipeline is proposed to quantify Aβ plaques in whole rat brain through several terabytes of image data acquired by a high-speed volumetric imaging approach we have developed previously. A novel segmentation framework applying a high-performance weakly supervised learning method which can dramatically reduce the human labeling consumption is described in this study. The effectiveness of our segmentation framework is validated with different metrics. The segmented Aβ plaques were mapped to a standard rat brain atlas for quantitative analysis of the Aβ distribution in each brain area. This pipeline may also be applied to the segmentation and accurate quantification of other non-specific morphology objects.
Collapse
Affiliation(s)
- Zhiyi Chen
- National Engineering Laboratory for Brain-Inspired Intelligence Technology and Application, School of Information Science and Technology, University of Science and Technology of China, Hefei, China,Institute of Artificial Intelligence, Hefei Comprehensive National Science Center, Hefei, China
| | - Weijie Zheng
- Institute of Artificial Intelligence, Hefei Comprehensive National Science Center, Hefei, China,AHU-IAI AI Joint Laboratory, Anhui University, Hefei, China
| | - Keliang Pang
- School of Pharmaceutical Sciences, IDG/McGovern Institute for Brain Research, Tsinghua University-Peking University Joint Center for Life Sciences, Tsinghua University, Beijing, China,*Correspondence: Keliang Pang,
| | - Debin Xia
- National Engineering Laboratory for Brain-Inspired Intelligence Technology and Application, School of Information Science and Technology, University of Science and Technology of China, Hefei, China,Institute of Artificial Intelligence, Hefei Comprehensive National Science Center, Hefei, China
| | - Lingxiao Guo
- Institute of Artificial Intelligence, Hefei Comprehensive National Science Center, Hefei, China
| | - Xuejin Chen
- National Engineering Laboratory for Brain-Inspired Intelligence Technology and Application, School of Information Science and Technology, University of Science and Technology of China, Hefei, China,Institute of Artificial Intelligence, Hefei Comprehensive National Science Center, Hefei, China
| | - Feng Wu
- National Engineering Laboratory for Brain-Inspired Intelligence Technology and Application, School of Information Science and Technology, University of Science and Technology of China, Hefei, China,Institute of Artificial Intelligence, Hefei Comprehensive National Science Center, Hefei, China
| | - Hao Wang
- National Engineering Laboratory for Brain-Inspired Intelligence Technology and Application, School of Information Science and Technology, University of Science and Technology of China, Hefei, China,Institute of Artificial Intelligence, Hefei Comprehensive National Science Center, Hefei, China,Hao Wang,
| |
Collapse
|
19
|
Giancardo L, Niktabe A, Ocasio L, Abdelkhaleq R, Salazar-Marioni S, Sheth SA. Segmentation of acute stroke infarct core using image-level labels on CT-angiography. Neuroimage Clin 2023; 37:103362. [PMID: 36893661 PMCID: PMC10011814 DOI: 10.1016/j.nicl.2023.103362] [Citation(s) in RCA: 1] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 08/01/2022] [Revised: 02/22/2023] [Accepted: 02/24/2023] [Indexed: 03/02/2023]
Abstract
Acute ischemic stroke is a leading cause of death and disability in the world. Treatment decisions, especially around emergent revascularization procedures, rely heavily on size and location of the infarct core. Currently, accurate assessment of this measure is challenging. While MRI-DWI is considered the gold standard, its availability is limited for most patients suffering from stroke. Another well-studied imaging modality is CT-Perfusion (CTP) which is much more common than MRI-DWI in acute stroke care, but not as precise as MRI-DWI, and it is still unavailable in many stroke hospitals. A method to determine infarct core using CT-Angiography (CTA), a much more available imaging modality albeit with significantly less contrast in stroke core area than CTP or MRI-DWI, would enable significantly better treatment decisions for stroke patients throughout the world. Existing deep-learning-based approaches for stroke core estimation have to face the trade-off between voxel-level segmentation / image-level labels and the difficulty of obtaining large enough samples of high-quality DWI images. The former occurs when algorithms can either output voxel-level labeling which is more informative but requires a significant effort by annotators, or image-level labels that allow for much simpler labeling of the images but results in less informative and interpretable output; the latter is a common issue that forces training either on small training sets using DWI as the target or larger, but noisier, dataset using CT-Perfusion (CTP) as the target. In this work, we present a deep learning approach including a new weighted gradient-based approach to obtain stroke core segmentation with image-level labeling, specifically the size of the acute stroke core volume. Additionally, this strategy allows us to train using labels derived from CTP estimations. We find that the proposed approach outperforms segmentation approaches trained on voxel-level data and the CTP estimation themselves.
Collapse
Affiliation(s)
- Luca Giancardo
- School of Biomedical Informatics, University of Texas Health Science Center at Houston, Houston, TX 77030 USA.
| | - Arash Niktabe
- McGovern Medical School, University of Texas Health Science Center at Houston, Houston, TX 77030 USA
| | - Laura Ocasio
- School of Biomedical Informatics, University of Texas Health Science Center at Houston, Houston, TX 77030 USA
| | - Rania Abdelkhaleq
- McGovern Medical School, University of Texas Health Science Center at Houston, Houston, TX 77030 USA
| | - Sergio Salazar-Marioni
- McGovern Medical School, University of Texas Health Science Center at Houston, Houston, TX 77030 USA
| | - Sunil A Sheth
- McGovern Medical School, University of Texas Health Science Center at Houston, Houston, TX 77030 USA
| |
Collapse
|
20
|
He Q, He L, Duan H, Sun Q, Zheng R, Guan J, He Y, Huang W, Guan T. Expression site agnostic histopathology image segmentation framework by self supervised domain adaption. Comput Biol Med 2023; 152:106412. [PMID: 36516576 DOI: 10.1016/j.compbiomed.2022.106412] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/28/2022] [Revised: 11/22/2022] [Accepted: 12/03/2022] [Indexed: 12/12/2022]
Abstract
MOTIVATION With the sites of antigen expression different, the segmentation of immunohistochemical (IHC) histopathology images is challenging, due to the visual variances. With H&E images highlighting the tissue structure and cell distribution more broadly, transferring more salient features from H&E images can achieve considerable performance on expression site agnostic IHC images segmentation. METHODS To the best of our knowledge, this is the first work that focuses on domain adaptive segmentation for different expression sites. We propose an expression site agnostic domain adaptive histopathology image semantic segmentation framework (ESASeg). In ESASeg, multi-level feature alignment encodes expression site invariance by learning generic representations of global and multi-scale local features. Moreover, self-supervision enhances domain adaptation to perceive high-level semantics by predicting pseudo-labels. RESULTS We construct a dataset with three IHCs (Her2 with membrane stained, Ki67 with nucleus stained, GPC3 with cytoplasm stained) with different expression sites from two diseases (breast and liver cancer). Intensive experiments on tumor region segmentation illustrate that ESASeg performs best across all metrics, and the implementation of each module proves to achieve impressive improvements. CONCLUSION The performance of ESASeg on the tumor region segmentation demonstrates the efficiency of the proposed framework, which provides a novel solution on expression site agnostic IHC related tasks. Moreover, the proposed domain adaption and self-supervision module can improve feature domain adaption and extraction without labels. In addition, ESASeg lays the foundation to perform joint analysis and information interaction for IHCs with different expression sites.
Collapse
Affiliation(s)
- Qiming He
- Institute of Biopharmaceutical and Health Engineering, Tsinghua Shenzhen International Graduate School, Shenzhen 518055, China.
| | - Ling He
- Institute of Biopharmaceutical and Health Engineering, Tsinghua Shenzhen International Graduate School, Shenzhen 518055, China.
| | - Hufei Duan
- Institute of Biopharmaceutical and Health Engineering, Tsinghua Shenzhen International Graduate School, Shenzhen 518055, China.
| | - Qiehe Sun
- Institute of Biopharmaceutical and Health Engineering, Tsinghua Shenzhen International Graduate School, Shenzhen 518055, China.
| | - Runliang Zheng
- Institute of Biopharmaceutical and Health Engineering, Tsinghua Shenzhen International Graduate School, Shenzhen 518055, China.
| | - Jian Guan
- National Cancer Center/National Clinical Research Center for Cancer/Cancer Hospital & Shenzhen Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Shenzhen, 518116, China.
| | - Yonghong He
- Institute of Biopharmaceutical and Health Engineering, Tsinghua Shenzhen International Graduate School, Shenzhen 518055, China.
| | - Wenting Huang
- National Cancer Center/National Clinical Research Center for Cancer/Cancer Hospital & Shenzhen Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Shenzhen, 518116, China.
| | - Tian Guan
- Institute of Biopharmaceutical and Health Engineering, Tsinghua Shenzhen International Graduate School, Shenzhen 518055, China.
| |
Collapse
|
21
|
Wang Z, Saoud C, Wangsiricharoen S, James AW, Popel AS, Sulam J. Label Cleaning Multiple Instance Learning: Refining Coarse Annotations on Single Whole-Slide Images. IEEE TRANSACTIONS ON MEDICAL IMAGING 2022; 41:3952-3968. [PMID: 36037454 PMCID: PMC9825360 DOI: 10.1109/tmi.2022.3202759] [Citation(s) in RCA: 8] [Impact Index Per Article: 4.0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Figures] [Subscribe] [Scholar Register] [Indexed: 06/15/2023]
Abstract
Annotating cancerous regions in whole-slide images (WSIs) of pathology samples plays a critical role in clinical diagnosis, biomedical research, and machine learning algorithms development. However, generating exhaustive and accurate annotations is labor-intensive, challenging, and costly. Drawing only coarse and approximate annotations is a much easier task, less costly, and it alleviates pathologists' workload. In this paper, we study the problem of refining these approximate annotations in digital pathology to obtain more accurate ones. Some previous works have explored obtaining machine learning models from these inaccurate annotations, but few of them tackle the refinement problem where the mislabeled regions should be explicitly identified and corrected, and all of them require a - often very large - number of training samples. We present a method, named Label Cleaning Multiple Instance Learning (LC-MIL), to refine coarse annotations on a single WSI without the need for external training data. Patches cropped from a WSI with inaccurate labels are processed jointly within a multiple instance learning framework, mitigating their impact on the predictive model and refining the segmentation. Our experiments on a heterogeneous WSI set with breast cancer lymph node metastasis, liver cancer, and colorectal cancer samples show that LC-MIL significantly refines the coarse annotations, outperforming state-of-the-art alternatives, even while learning from a single slide. Moreover, we demonstrate how real annotations drawn by pathologists can be efficiently refined and improved by the proposed approach. All these results demonstrate that LC-MIL is a promising, lightweight tool to provide fine-grained annotations from coarsely annotated pathology sets.
Collapse
|
22
|
Bateson M, Kervadec H, Dolz J, Lombaert H, Ben Ayed I. Source-free domain adaptation for image segmentation. Med Image Anal 2022; 82:102617. [DOI: 10.1016/j.media.2022.102617] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/21/2021] [Revised: 07/25/2022] [Accepted: 09/02/2022] [Indexed: 11/25/2022]
|
23
|
Del Amor R, Meseguer P, Parigi TL, Villanacci V, Colomer A, Launet L, Bazarova A, Tontini GE, Bisschops R, de Hertogh G, Ferraz JG, Götz M, Gui X, Hayee B, Lazarev M, Panaccione R, Parra-Blanco A, Bhandari P, Pastorelli L, Rath T, Røyset ES, Vieth M, Zardo D, Grisan E, Ghosh S, Iacucci M, Naranjo V. Constrained multiple instance learning for ulcerative colitis prediction using histological images. COMPUTER METHODS AND PROGRAMS IN BIOMEDICINE 2022; 224:107012. [PMID: 35843078 DOI: 10.1016/j.cmpb.2022.107012] [Citation(s) in RCA: 10] [Impact Index Per Article: 5.0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Subscribe] [Scholar Register] [Received: 05/06/2022] [Revised: 07/05/2022] [Accepted: 07/06/2022] [Indexed: 06/15/2023]
Abstract
BACKGROUND AND OBJECTIVE Ulcerative colitis (UC) is an inflammatory bowel disease (IBD) affecting the colon and the rectum characterized by a remitting-relapsing course. To detect mucosal inflammation associated with UC, histology is considered the most stringent criteria. In turn, histologic remission (HR) correlates with improved clinical outcomes and has been recently recognized as a desirable treatment target. The leading biomarker for assessing histologic remission is the presence or absence of neutrophils. Therefore, the finding of this cell in specific colon structures indicates that the patient has UC activity. However, no previous studies based on deep learning have been developed to identify UC based on neutrophils detection using whole-slide images (WSI). METHODS The methodological core of this work is a novel multiple instance learning (MIL) framework with location constraints able to determine the presence of UC activity using WSI. In particular, we put forward an effective way to introduce constraints about positive instances to effectively explore additional weakly supervised information that is easy to obtain and enjoy a significant boost to the learning process. In addition, we propose a new weighted embedding to enlarge the relevance of the positive instances. RESULTS Extensive experiments on a multi-center dataset of colon and rectum WSIs, PICASSO-MIL, demonstrate that using the location information we can improve considerably the results at WSI-level. In comparison with prior MIL settings, our method allows for 10% improvements in bag-level accuracy. CONCLUSION Our model, which introduces a new form of constraints, surpass the results achieved from current state-of-the-art methods that focus on the MIL paradigm. Our method can be applied to other histological concerns where the morphological features determining a positive WSI are tiny and similar to others in the image.
Collapse
Affiliation(s)
- Rocío Del Amor
- Instituto de Investigación e Innovación en Bioingeniería, Universitat Politécnica de Valéncia, Valencia, Spain.
| | - Pablo Meseguer
- Instituto de Investigación e Innovación en Bioingeniería, Universitat Politécnica de Valéncia, Valencia, Spain
| | - Tommaso Lorenzo Parigi
- Department of Biomedical Sciences, Humanitas University, Milan, Italy; University of Birmingham, Immunology and Immunotherapy, Birmingham, United Kingdom
| | - Vincenzo Villanacci
- Institute of Pathology, ASST Spedali Civili, University of Brescia, Brescia, Italy
| | - Adrián Colomer
- Instituto de Investigación e Innovación en Bioingeniería, Universitat Politécnica de Valéncia, Valencia, Spain
| | - Laëtitia Launet
- Instituto de Investigación e Innovación en Bioingeniería, Universitat Politécnica de Valéncia, Valencia, Spain
| | - Alina Bazarova
- Institute for Biological Physics, University of Cologne, Cologne, Germany
| | - Gian Eugenio Tontini
- Fondazione IRCCS Ca'Granda Ospedale Maggiore Policlinico, Department of Pathophysiology and Transplantation, University of Milan, Milan, Italy
| | - Raf Bisschops
- Division of Gastroenterology, University Hospitals Leuven, Leuven, Belgium
| | - Gert de Hertogh
- Division of Gastroenterology, University Hospitals Leuven, Leuven, Belgium
| | - Jose G Ferraz
- Division of Gastroenterology, University of Calgary Cumming School of Medicine, Calgary, Canada
| | - Martin Götz
- Division of Gastroenterology, Klinikum, Böblingen, Germany
| | - Xianyong Gui
- Department of Laboratory Medicine and Pathology, University of Washington, Seattle, USA
| | - Bu'Hussain Hayee
- Division of Gastroenterology, Kings College London, London, United Kingdom
| | - Mark Lazarev
- Division of Gastroenterology, Johns Hopkins Hospital, Baltimore, United States
| | - Remo Panaccione
- Division of Gastroenterology, University of Calgary Cumming School of Medicine, Calgary, Canada
| | - Adolfo Parra-Blanco
- Division of Gastroenterology, University of Nottingham, Nottingham, United Kingdom
| | - Pradeep Bhandari
- Division of Gastroenterology, Queen Alexandra Hospital, Portsmouth, United Kingdom
| | - Luca Pastorelli
- Liver and Gastroenterology Unit, Universita' degli Studi di Milano, ASST Santi Paolo E Carlo, University Hospital San Paolo, Milan, Italy
| | - Timo Rath
- Division of Gastroenterology, University of Erlangen, Erlangen, Germany
| | - Elin Synnøve Røyset
- Department of Clinical and Molecular Medicine, Norwegian University of Science and Technology, Trondheim, Norway
| | - Michael Vieth
- Klinikum Bayreuth, Bayreuth, Germany; Institute of Pathology, Friedrich-Alexander-University Erlangen-Nuremberg, Nuremberg, Germany
| | - Davide Zardo
- Department of Pathology, San Bortolo Hospital, Vicenza, Italy
| | - Enrico Grisan
- Department of Information Engineering, Padova, Italy; School of Engineering, London South Bank University, London, UK
| | - Subrata Ghosh
- APC Microbiome Ireland, College of Medicine and Health, Cork, Ireland; Institute for Biological Physics, University of Cologne, Cologne, Germany
| | - Marietta Iacucci
- Department of Biomedical Sciences, Humanitas University, Milan, Italy; National Institute for Health Research (NIHR) Biomedical Research Centre, Birmingham, United Kingdom; Department of Gastroenterology, University Hospitals Birmingham NHS Trust, Birmingham, United Kingdom
| | - Valery Naranjo
- Instituto de Investigación e Innovación en Bioingeniería, Universitat Politécnica de Valéncia, Valencia, Spain
| |
Collapse
|
24
|
Wu H, Pang KKY, Pang GKH, Au-Yeung RKH. A soft-computing based approach to overlapped cells analysis in histopathology images with genetic algorithm. Appl Soft Comput 2022. [DOI: 10.1016/j.asoc.2022.109279] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/30/2022]
|
25
|
Liu Y, He Q, Duan H, Shi H, Han A, He Y. Using Sparse Patch Annotation for Tumor Segmentation in Histopathological Images. SENSORS (BASEL, SWITZERLAND) 2022; 22:6053. [PMID: 36015814 PMCID: PMC9414209 DOI: 10.3390/s22166053] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Figures] [Subscribe] [Scholar Register] [Received: 07/19/2022] [Revised: 08/05/2022] [Accepted: 08/10/2022] [Indexed: 06/15/2023]
Abstract
Tumor segmentation is a fundamental task in histopathological image analysis. Creating accurate pixel-wise annotations for such segmentation tasks in a fully-supervised training framework requires significant effort. To reduce the burden of manual annotation, we propose a novel weakly supervised segmentation framework based on sparse patch annotation, i.e., only small portions of patches in an image are labeled as 'tumor' or 'normal'. The framework consists of a patch-wise segmentation model called PSeger, and an innovative semi-supervised algorithm. PSeger has two branches for patch classification and image classification, respectively. This two-branch structure enables the model to learn more general features and thus reduce the risk of overfitting when learning sparsely annotated data. We incorporate the idea of consistency learning and self-training into the semi-supervised training strategy to take advantage of the unlabeled images. Trained on the BCSS dataset with only 25% of the images labeled (five patches for each labeled image), our proposed method achieved competitive performance compared to the fully supervised pixel-wise segmentation models. Experiments demonstrate that the proposed solution has the potential to reduce the burden of labeling histopathological images.
Collapse
Affiliation(s)
- Yiqing Liu
- Institute of Biopharmaceutical and Health Engineering, Tsinghua Shenzhen International Graduate School, Shenzhen 518055, China
| | - Qiming He
- Institute of Biopharmaceutical and Health Engineering, Tsinghua Shenzhen International Graduate School, Shenzhen 518055, China
| | - Hufei Duan
- Institute of Biopharmaceutical and Health Engineering, Tsinghua Shenzhen International Graduate School, Shenzhen 518055, China
| | - Huijuan Shi
- Department of Pathology, The First Affiliated Hospital, Sun Yat-sen University, Guangzhou 510080, China
| | - Anjia Han
- Department of Pathology, The First Affiliated Hospital, Sun Yat-sen University, Guangzhou 510080, China
| | - Yonghong He
- Institute of Biopharmaceutical and Health Engineering, Tsinghua Shenzhen International Graduate School, Shenzhen 518055, China
| |
Collapse
|
26
|
Silva-Rodríguez J, Naranjo V, Dolz J. Constrained unsupervised anomaly segmentation. Med Image Anal 2022; 80:102526. [DOI: 10.1016/j.media.2022.102526] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 02/16/2022] [Revised: 05/29/2022] [Accepted: 06/24/2022] [Indexed: 10/17/2022]
|
27
|
Wang K, Wang Y, Zhan B, Yang Y, Zu C, Wu X, Zhou J, Nie D, Zhou L. An Efficient Semi-Supervised Framework with Multi-Task and Curriculum Learning for Medical Image Segmentation. Int J Neural Syst 2022; 32:2250043. [DOI: 10.1142/s0129065722500435] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.5] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/18/2022]
|
28
|
Adiga V S, Dolz J, Lombaert H. Attention-based Dynamic Subspace Learners for Medical Image Analysis. IEEE J Biomed Health Inform 2022; 26:4599-4610. [PMID: 35763468 DOI: 10.1109/jbhi.2022.3186882] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/09/2022]
Abstract
Learning similarity is a key aspect in medical image analysis, particularly in recommendation systems or in uncovering the interpretation of anatomical data in images. Most existing methods learn such similarities in the embedding space over image sets using a single metric learner. Images, however, have a variety of object attributes such as color, shape, or artifacts. Encoding such attributes using a single metric learner is inadequate and may fail to generalize. Instead, multiple learners could focus on separate aspects of these attributes in subspaces of an overarching embedding. This, however, implies the number of learners to be found empirically for each new dataset. This work, Dynamic Subspace Learners, proposes to dynamically exploit multiple learners by removing the need of knowing apriori the number of learners and aggregating new subspace learners during training. Furthermore, the visual interpretability of such subspace learning is enforced by integrating an attention module into our method. This integrated attention mechanism provides a visual insight of discriminative image features that contribute to the clustering of image sets and a visual explanation of the embedding features. The benefits of our attention-based dynamic subspace learners are evaluated in the application of image clustering, image retrieval, and weakly supervised segmentation. Our method achieves competitive results with the performances of multiple learners baselines and significantly outperforms the classification network in terms of clustering and retrieval scores on three different public benchmark datasets. Moreover, our method also provides an attention map generated directly during inference to illustrate the visual interpretability of the embedding features. These attention maps offer a proxy-labels, which improves the segmentation accuracy up to 15% in Dice scores when compared to state-of-the-art interpretation techniques.
Collapse
|
29
|
Silva-Rodríguez J, Schmidt A, Sales MA, Molina R, Naranjo V. Proportion constrained weakly supervised histopathology image classification. Comput Biol Med 2022; 147:105714. [PMID: 35753089 DOI: 10.1016/j.compbiomed.2022.105714] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 02/07/2022] [Revised: 05/14/2022] [Accepted: 06/04/2022] [Indexed: 11/16/2022]
Abstract
Multiple instance learning (MIL) deals with data grouped into bags of instances, of which only the global information is known. In recent years, this weakly supervised learning paradigm has become very popular in histological image analysis because it alleviates the burden of labeling all cancerous regions of large Whole Slide Images (WSIs) in detail. However, these methods require large datasets to perform properly, and many approaches only focus on simple binary classification. This often does not match the real-world problems where multi-label settings are frequent and possible constraints must be taken into account. In this work, we propose a novel multi-label MIL formulation based on inequality constraints that is able to incorporate prior knowledge about instance proportions. Our method has a theoretical foundation in optimization with log-barrier extensions, applied to bag-level class proportions. This encourages the model to respect the proportion ordering during training. Extensive experiments on a new public dataset of prostate cancer WSIs analysis, SICAP-MIL, demonstrate that using the prior proportion information we can achieve instance-level results similar to supervised methods on datasets of similar size. In comparison with prior MIL settings, our method allows for ∼13% improvements in instance-level accuracy, and ∼3% in the multi-label mean area under the ROC curve at the bag-level.
Collapse
Affiliation(s)
- Julio Silva-Rodríguez
- Institute of Transport and Territory, Universitat Politècnica de València, Valencia, Spain.
| | - Arne Schmidt
- Department of Computer Science and Artificial Intelligence, University of Granada, Granada, Spain.
| | - María A Sales
- Anatomical Pathology Service, University Clinical Hospital of Valencia, Valencia, Spain.
| | - Rafael Molina
- Department of Computer Science and Artificial Intelligence, University of Granada, Granada, Spain.
| | - Valery Naranjo
- Institute of Research and Innovation in Bioengineering, Universitat Politècnica de València, Valencia, Spain.
| |
Collapse
|
30
|
Multi-layer pseudo-supervision for histopathology tissue semantic segmentation using patch-level classification labels. Med Image Anal 2022; 80:102487. [PMID: 35671591 DOI: 10.1016/j.media.2022.102487] [Citation(s) in RCA: 12] [Impact Index Per Article: 6.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/06/2021] [Revised: 05/07/2022] [Accepted: 05/20/2022] [Indexed: 01/15/2023]
Abstract
Tissue-level semantic segmentation is a vital step in computational pathology. Fully-supervised models have already achieved outstanding performance with dense pixel-level annotations. However, drawing such labels on the giga-pixel whole slide images is extremely expensive and time-consuming. In this paper, we use only patch-level classification labels to achieve tissue semantic segmentation on histopathology images, finally reducing the annotation efforts. We propose a two-step model including a classification and a segmentation phases. In the classification phase, we propose a CAM-based model to generate pseudo masks by patch-level labels. In the segmentation phase, we achieve tissue semantic segmentation by our propose Multi-Layer Pseudo-Supervision. Several technical novelties have been proposed to reduce the information gap between pixel-level and patch-level annotations. As a part of this paper, we introduce a new weakly-supervised semantic segmentation (WSSS) dataset for lung adenocarcinoma (LUAD-HistoSeg). We conduct several experiments to evaluate our proposed model on two datasets. Our proposed model outperforms five state-of-the-art WSSS approaches. Note that we can achieve comparable quantitative and qualitative results with the fully-supervised model, with only around a 2% gap for MIoU and FwIoU. By comparing with manual labeling on a randomly sampled 100 patches dataset, patch-level labeling can greatly reduce the annotation time from hours to minutes. The source code and the released datasets are available at: https://github.com/ChuHan89/WSSS-Tissue.
Collapse
|
31
|
Lyu F, Ma AJ, Yip TCF, Wong GLH, Yuen PC. Weakly Supervised Liver Tumor Segmentation Using Couinaud Segment Annotation. IEEE TRANSACTIONS ON MEDICAL IMAGING 2022; 41:1138-1149. [PMID: 34871168 DOI: 10.1109/tmi.2021.3132905] [Citation(s) in RCA: 3] [Impact Index Per Article: 1.5] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 06/13/2023]
Abstract
Automatic liver tumor segmentation is of great importance for assisting doctors in liver cancer diagnosis and treatment planning. Recently, deep learning approaches trained with pixel-level annotations have contributed many breakthroughs in image segmentation. However, acquiring such accurate dense annotations is time-consuming and labor-intensive, which limits the performance of deep neural networks for medical image segmentation. We note that Couinaud segment is widely used by radiologists when recording liver cancer-related findings in the reports, since it is well-suited for describing the localization of tumors. In this paper, we propose a novel approach to train convolutional networks for liver tumor segmentation using Couinaud segment annotations. Couinaud segment annotations are image-level labels with values ranging from 1 to 8, indicating a specific region of the liver. Our proposed model, namely CouinaudNet, can estimate pseudo tumor masks from the Couinaud segment annotations as pixel-wise supervision for training a fully supervised tumor segmentation model, and it is composed of two components: 1) an inpainting network with Couinaud segment masks which can effectively remove tumors for pathological images by filling the tumor regions with plausible healthy-looking intensities; 2) a difference spotting network for segmenting the tumors, which is trained with healthy-pathological pairs generated by an effective tumor synthesis strategy. The proposed method is extensively evaluated on two liver tumor segmentation datasets. The experimental results demonstrate that our method can achieve competitive performance compared to the fully supervised counterpart and the state-of-the-art methods while requiring significantly less annotation effort.
Collapse
|
32
|
Weakly Supervised Segmentation on Neural Compressed Histopathology with Self-Equivariant Regularization. Med Image Anal 2022; 80:102482. [DOI: 10.1016/j.media.2022.102482] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/22/2021] [Revised: 04/06/2022] [Accepted: 05/20/2022] [Indexed: 11/17/2022]
|
33
|
Kiran I, Raza B, Ijaz A, Khan MA. DenseRes-Unet: Segmentation of overlapped/clustered nuclei from multi organ histopathology images. Comput Biol Med 2022; 143:105267. [PMID: 35114445 DOI: 10.1016/j.compbiomed.2022.105267] [Citation(s) in RCA: 16] [Impact Index Per Article: 8.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 10/08/2021] [Revised: 01/23/2022] [Accepted: 01/23/2022] [Indexed: 11/16/2022]
Abstract
Cancer is the second deadliest disease globally that can affect any human body organ. Early detection of cancer can increase the chances of survival in humans. Morphometric appearances of histopathology images make it difficult to segment nuclei effectively. We proposed a model to segment overlapped nuclei from H&E stained images. U-Net model achieved state-of-the-art performance in many medical image segmentation tasks; however, we modified the U-Net to learn a distinct set of consistent features. In this paper, we proposed the DenseRes-Unet model by integrating dense blocks in the last layers of the encoder block of U-Net, focused on relevant features from previous layers of the model. Moreover, we take advantage of residual connections with Atrous blocks instead of conventional skip connections, which helps to reduce the semantic gap between encoder and decoder paths. The distance map and binary threshold techniques intensify the nuclei interior and contour information in the images, respectively. The distance map is used to detect the center point of nuclei; moreover, it differentiates among nuclei interior boundary and core area. The distance map lacks a contour problem, which is resolved by using a binary threshold. Binary threshold helps to enhance the pixels around nuclei. Afterward, we fed images into the proposed DenseRes-Unet model, a deep, fully convolutional network to segment nuclei in the images. We have evaluated our model on four publicly available datasets for Nuclei segmentation to validate the model's performance. Our proposed model achieves 89.77% accuracy 90.36% F1-score, and 78.61% Aggregated Jaccard Index (AJI) on Multi organ Nucleus Segmentation (MoNuSeg).
Collapse
Affiliation(s)
- Iqra Kiran
- Department of Computer Science, COMSATS University Islamabad (CUI), Islamabad, Pakistan.
| | - Basit Raza
- Department of Computer Science, COMSATS University Islamabad (CUI), Islamabad, Pakistan.
| | - Areesha Ijaz
- Department of Computer Science, COMSATS University Islamabad (CUI), Islamabad, Pakistan.
| | - Muazzam A Khan
- Department of Computer Science, Quaid-i-Azam University, Islamabad, Pakistan.
| |
Collapse
|
34
|
Thiagarajan P, Khairnar P, Ghosh S. Explanation and Use of Uncertainty Quantified by Bayesian Neural Network Classifiers for Breast Histopathology Images. IEEE TRANSACTIONS ON MEDICAL IMAGING 2022; 41:815-825. [PMID: 34699354 DOI: 10.1109/tmi.2021.3123300] [Citation(s) in RCA: 5] [Impact Index Per Article: 2.5] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 06/13/2023]
Abstract
Despite the promise of Convolutional neural network (CNN) based classification models for histopathological images, it is infeasible to quantify its uncertainties. Moreover, CNNs may suffer from overfitting when the data is biased. We show that Bayesian-CNN can overcome these limitations by regularizing automatically and by quantifying the uncertainty. We have developed a novel technique to utilize the uncertainties provided by the Bayesian-CNN that significantly improves the performance on a large fraction of the test data (about 6% improvement in accuracy on 77% of test data). Further, we provide a novel explanation for the uncertainty by projecting the data into a low dimensional space through a nonlinear dimensionality reduction technique. This dimensionality reduction enables interpretation of the test data through visualization and reveals the structure of the data in a low dimensional feature space. We show that the Bayesian-CNN can perform much better than the state-of-the-art transfer learning CNN (TL-CNN) by reducing the false negative and false positive by 11% and 7.7% respectively for the present data set. It achieves this performance with only 1.86 million parameters as compared to 134.33 million for TL-CNN. Besides, we modify the Bayesian-CNN by introducing a stochastic adaptive activation function. The modified Bayesian-CNN performs slightly better than Bayesian-CNN on all performance metrics and significantly reduces the number of false negatives and false positives (3% reduction for both). We also show that these results are statistically significant by performing McNemar's statistical significance test. This work shows the advantages of Bayesian-CNN against the state-of-the-art, explains and utilizes the uncertainties for histopathological images. It should find applications in various medical image classifications.
Collapse
|
35
|
Chen RJ, Lu MY, Wang J, Williamson DFK, Rodig SJ, Lindeman NI, Mahmood F. Pathomic Fusion: An Integrated Framework for Fusing Histopathology and Genomic Features for Cancer Diagnosis and Prognosis. IEEE TRANSACTIONS ON MEDICAL IMAGING 2022; 41:757-770. [PMID: 32881682 DOI: 10.1109/tmi.2020.3021387] [Citation(s) in RCA: 116] [Impact Index Per Article: 58.0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Grants] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 05/23/2023]
Abstract
Cancer diagnosis, prognosis, mymargin and therapeutic response predictions are based on morphological information from histology slides and molecular profiles from genomic data. However, most deep learning-based objective outcome prediction and grading paradigms are based on histology or genomics alone and do not make use of the complementary information in an intuitive manner. In this work, we propose Pathomic Fusion, an interpretable strategy for end-to-end multimodal fusion of histology image and genomic (mutations, CNV, RNA-Seq) features for survival outcome prediction. Our approach models pairwise feature interactions across modalities by taking the Kronecker product of unimodal feature representations, and controls the expressiveness of each representation via a gating-based attention mechanism. Following supervised learning, we are able to interpret and saliently localize features across each modality, and understand how feature importance shifts when conditioning on multimodal input. We validate our approach using glioma and clear cell renal cell carcinoma datasets from the Cancer Genome Atlas (TCGA), which contains paired whole-slide image, genotype, and transcriptome data with ground truth survival and histologic grade labels. In a 15-fold cross-validation, our results demonstrate that the proposed multimodal fusion paradigm improves prognostic determinations from ground truth grading and molecular subtyping, as well as unimodal deep networks trained on histology and genomic data alone. The proposed method establishes insight and theory on how to train deep networks on multimodal biomedical data in an intuitive manner, which will be useful for other problems in medicine that seek to combine heterogeneous data streams for understanding diseases and predicting response and resistance to treatment. Code and trained models are made available at: https://github.com/mahmoodlab/PathomicFusion.
Collapse
|
36
|
Deep Learning on Histopathological Images for Colorectal Cancer Diagnosis: A Systematic Review. Diagnostics (Basel) 2022; 12:diagnostics12040837. [PMID: 35453885 PMCID: PMC9028395 DOI: 10.3390/diagnostics12040837] [Citation(s) in RCA: 17] [Impact Index Per Article: 8.5] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 02/28/2022] [Revised: 03/22/2022] [Accepted: 03/25/2022] [Indexed: 02/04/2023] Open
Abstract
Colorectal cancer (CRC) is the second most common cancer in women and the third most common in men, with an increasing incidence. Pathology diagnosis complemented with prognostic and predictive biomarker information is the first step for personalized treatment. The increased diagnostic load in the pathology laboratory, combined with the reported intra- and inter-variability in the assessment of biomarkers, has prompted the quest for reliable machine-based methods to be incorporated into the routine practice. Recently, Artificial Intelligence (AI) has made significant progress in the medical field, showing potential for clinical applications. Herein, we aim to systematically review the current research on AI in CRC image analysis. In histopathology, algorithms based on Deep Learning (DL) have the potential to assist in diagnosis, predict clinically relevant molecular phenotypes and microsatellite instability, identify histological features related to prognosis and correlated to metastasis, and assess the specific components of the tumor microenvironment.
Collapse
|
37
|
Weakly-supervised learning for catheter segmentation in 3D frustum ultrasound. Comput Med Imaging Graph 2022; 96:102037. [DOI: 10.1016/j.compmedimag.2022.102037] [Citation(s) in RCA: 2] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 12/18/2020] [Revised: 11/15/2021] [Accepted: 01/13/2022] [Indexed: 11/21/2022]
|
38
|
Belharbi S, Rony J, Dolz J, Ayed IB, Mccaffrey L, Granger E. Deep Interpretable Classification and Weakly-Supervised Segmentation of Histology Images via Max-Min Uncertainty. IEEE TRANSACTIONS ON MEDICAL IMAGING 2022; 41:702-714. [PMID: 34705638 DOI: 10.1109/tmi.2021.3123461] [Citation(s) in RCA: 3] [Impact Index Per Article: 1.5] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 06/13/2023]
Abstract
Weakly-supervised learning (WSL) has recently triggered substantial interest as it mitigates the lack of pixel-wise annotations. Given global image labels, WSL methods yield pixel-level predictions (segmentations), which enable to interpret class predictions. Despite their recent success, mostly with natural images, such methods can face important challenges when the foreground and background regions have similar visual cues, yielding high false-positive rates in segmentations, as is the case in challenging histology images. WSL training is commonly driven by standard classification losses, which implicitly maximize model confidence, and locate the discriminative regions linked to classification decisions. Therefore, they lack mechanisms for modeling explicitly non-discriminative regions and reducing false-positive rates. We propose novel regularization terms, which enable the model to seek both non-discriminative and discriminative regions, while discouraging unbalanced segmentations. We introduce high uncertainty as a criterion to localize non-discriminative regions that do not affect classifier decision, and describe it with original Kullback-Leibler (KL) divergence losses evaluating the deviation of posterior predictions from the uniform distribution. Our KL terms encourage high uncertainty of the model when the latter inputs the latent non-discriminative regions. Our loss integrates: (i) a cross-entropy seeking a foreground, where model confidence about class prediction is high; (ii) a KL regularizer seeking a background, where model uncertainty is high; and (iii) log-barrier terms discouraging unbalanced segmentations. Comprehensive experiments and ablation studies over the public GlaS colon cancer data and a Camelyon16 patch-based benchmark for breast cancer show substantial improvements over state-of-the-art WSL methods, and confirm the effect of our new regularizers (our code is publicly available at https://github.com/sbelharbi/deep-wsl-histo-min-max-uncertainty).
Collapse
|
39
|
McGenity C, Wright A, Treanor D. AIM in Surgical Pathology. Artif Intell Med 2022. [DOI: 10.1007/978-3-030-64573-1_278] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/28/2022]
|
40
|
Patel G, Dolz J. Weakly supervised segmentation with cross-modality equivariant constraints. Med Image Anal 2022; 77:102374. [DOI: 10.1016/j.media.2022.102374] [Citation(s) in RCA: 2] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 04/19/2021] [Revised: 01/12/2022] [Accepted: 01/18/2022] [Indexed: 10/19/2022]
|
41
|
Zheng Y, Jiang Z, Shi J, Xie F, Zhang H, Luo W, Hu D, Sun S, Jiang Z, Xue C. Encoding histopathology whole slide images with location-aware graphs for diagnostically relevant regions retrieval. Med Image Anal 2021; 76:102308. [PMID: 34856455 DOI: 10.1016/j.media.2021.102308] [Citation(s) in RCA: 5] [Impact Index Per Article: 1.7] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 10/19/2019] [Revised: 10/14/2021] [Accepted: 11/17/2021] [Indexed: 01/18/2023]
Abstract
Content-based histopathological image retrieval (CBHIR) has become popular in recent years in histopathological image analysis. CBHIR systems provide auxiliary diagnosis information for pathologists by searching for and returning regions that are contently similar to the region of interest (ROI) from a pre-established database. It is challenging and yet significant in clinical applications to retrieve diagnostically relevant regions from a database consisting of histopathological whole slide images (WSIs). In this paper, we propose a novel framework for regions retrieval from WSI database based on location-aware graphs and deep hash techniques. Compared to the present CBHIR framework, both structural information and global location information of ROIs in the WSI are preserved by graph convolution and self-attention operations, which makes the retrieval framework more sensitive to regions that are similar in tissue distribution. Moreover, benefited from the graph structure, the proposed framework has good scalability for both the size and shape variation of ROIs. It allows the pathologist to define query regions using free curves according to the appearance of tissue. Thirdly, the retrieval is achieved based on the hash technique, which ensures the framework is efficient and adequate for practical large-scale WSI database. The proposed method was evaluated on an in-house endometrium dataset with 2650 WSIs and the public ACDC-LungHP dataset. The experimental results have demonstrated that the proposed method achieved a mean average precision above 0.667 on the endometrium dataset and above 0.869 on the ACDC-LungHP dataset in the task of irregular region retrieval, which are superior to the state-of-the-art methods. The average retrieval time from a database containing 1855 WSIs is 0.752 ms. The source code is available at https://github.com/zhengyushan/lagenet.
Collapse
Affiliation(s)
- Yushan Zheng
- Beijing Advanced Innovation Center for Biomedical Engineering, Beihang University, Beijing 100191, China
| | - Zhiguo Jiang
- Beijing Advanced Innovation Center for Biomedical Engineering, Beihang University, Beijing 100191, China; Image Processing Center, School of Astronautics, Beihang University, Beijing 102206, China.
| | - Jun Shi
- School of Software, Hefei University of Technology, Hefei 230601, China.
| | - Fengying Xie
- Beijing Advanced Innovation Center for Biomedical Engineering, Beihang University, Beijing 100191, China; Image Processing Center, School of Astronautics, Beihang University, Beijing 102206, China
| | - Haopeng Zhang
- Beijing Advanced Innovation Center for Biomedical Engineering, Beihang University, Beijing 100191, China; Image Processing Center, School of Astronautics, Beihang University, Beijing 102206, China
| | - Wei Luo
- Beijing Advanced Innovation Center for Biomedical Engineering, Beihang University, Beijing 100191, China; Image Processing Center, School of Astronautics, Beihang University, Beijing 102206, China
| | - Dingyi Hu
- Beijing Advanced Innovation Center for Biomedical Engineering, Beihang University, Beijing 100191, China; Image Processing Center, School of Astronautics, Beihang University, Beijing 102206, China
| | - Shujiao Sun
- Beijing Advanced Innovation Center for Biomedical Engineering, Beihang University, Beijing 100191, China; Image Processing Center, School of Astronautics, Beihang University, Beijing 102206, China
| | - Zhongmin Jiang
- Department of Pathology, Tianjin Fifth Central Hospital, Tianjin 300450, China
| | - Chenghai Xue
- Wankangyuan Tianjin Gene Technology, Inc, Tianjin 300220, China; Tianjin Institute of Industrial Biotechnology, Chinese Academy of Sciences, Tianjin 300308, China
| |
Collapse
|
42
|
Fu Q, Zhang Y, Wang P, Pi J, Qiu X, Guo Z, Huang Y, Zhao Y, Li S, Xu J. Rapid identification of the resistance of urinary tract pathogenic bacteria using deep learning-based spectroscopic analysis. Anal Bioanal Chem 2021; 413:7401-7410. [PMID: 34673992 DOI: 10.1007/s00216-021-03691-z] [Citation(s) in RCA: 8] [Impact Index Per Article: 2.7] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/24/2021] [Revised: 09/23/2021] [Accepted: 09/23/2021] [Indexed: 11/24/2022]
Abstract
The resistance of urinary tract pathogenic bacteria to various antibiotics is increasing, which requires the rapid detection of infectious pathogens for accurate and timely antibiotic treatment. Here, we propose a rapid diagnosis strategy for the antibiotic resistance of bacteria in urinary tract infections (UTIs) based on surface-enhanced Raman scattering (SERS) using a positively charged gold nanoparticle planar solid SERS substrate. Then, an intelligent identification model for SERS spectra based on the deep learning technique is constructed to realize the rapid, ultrasensitive, and non-labeled detection of pathogenic bacteria. A total of 54,000 SERS spectra were collected from 18 isolates belonging to 6 species of common UTI bacteria in this work to realize identification of bacterial species, antibiotic sensitivity, and multidrug resistance (MDR) via convolutional neural networks (CNN). This method significantly simplify the Raman data processing processes without background removing and smoothing, however, achieving 96% above classification accuracy, which was significantly greater than the 85% accuracy of the traditional multivariate statistical analysis algorithm principal component analysis combined with the K-nearest neighbor (PCA-KNN). This work clearly elucidated the potential of combining SERS and deep learning technique to realize culture-free identification of pathogenic bacteria and their associated antibiotic sensitivity.
Collapse
Affiliation(s)
- Qiuyue Fu
- Biomedical Photonics Laboratory, School of Biomedical Engineering, Guangdong Medical University, Dongguan, 523808, Guangdong, China
| | - Yanjiao Zhang
- School of Basic Medicine, Guangdong Medical University, Dongguan, 523808, China
| | - Peng Wang
- Biomedical Photonics Laboratory, School of Biomedical Engineering, Guangdong Medical University, Dongguan, 523808, Guangdong, China
| | - Jiang Pi
- Guangdong Provincial Key Laboratory of Medical Molecular Diagnostics, Guangdong Medical University, Dongguan, 523808, Guangdong, China
| | - Xun Qiu
- Biomedical Photonics Laboratory, School of Biomedical Engineering, Guangdong Medical University, Dongguan, 523808, Guangdong, China
| | - Zhusheng Guo
- Donghua Hospital Laboratory Department, Dongguan, 523808, Guangdong, China
| | - Ya Huang
- Donghua Hospital Laboratory Department, Dongguan, 523808, Guangdong, China
| | - Yi Zhao
- Guangdong Provincial Key Laboratory of Molecular Diagnosis, Guangdong Medical University, Dongguan, 523808, Guangdong, China
| | - Shaoxin Li
- Biomedical Photonics Laboratory, School of Biomedical Engineering, Guangdong Medical University, Dongguan, 523808, Guangdong, China.
| | - Junfa Xu
- Guangdong Provincial Key Laboratory of Medical Molecular Diagnostics, Guangdong Medical University, Dongguan, 523808, Guangdong, China.
| |
Collapse
|
43
|
Dang VN, Galati F, Cortese R, Di Giacomo G, Marconetto V, Mathur P, Lekadir K, Lorenzi M, Prados F, Zuluaga MA. Vessel-CAPTCHA: An efficient learning framework for vessel annotation and segmentation. Med Image Anal 2021; 75:102263. [PMID: 34731770 DOI: 10.1016/j.media.2021.102263] [Citation(s) in RCA: 9] [Impact Index Per Article: 3.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 01/29/2021] [Revised: 10/04/2021] [Accepted: 10/06/2021] [Indexed: 12/22/2022]
Abstract
Deep learning techniques for 3D brain vessel image segmentation have not been as successful as in the segmentation of other organs and tissues. This can be explained by two factors. First, deep learning techniques tend to show poor performances at the segmentation of relatively small objects compared to the size of the full image. Second, due to the complexity of vascular trees and the small size of vessels, it is challenging to obtain the amount of annotated training data typically needed by deep learning methods. To address these problems, we propose a novel annotation-efficient deep learning vessel segmentation framework. The framework avoids pixel-wise annotations, only requiring weak patch-level labels to discriminate between vessel and non-vessel 2D patches in the training set, in a setup similar to the CAPTCHAs used to differentiate humans from bots in web applications. The user-provided weak annotations are used for two tasks: (1) to synthesize pixel-wise pseudo-labels for vessels and background in each patch, which are used to train a segmentation network, and (2) to train a classifier network. The classifier network allows to generate additional weak patch labels, further reducing the annotation burden, and it acts as a second opinion for poor quality images. We use this framework for the segmentation of the cerebrovascular tree in Time-of-Flight angiography (TOF) and Susceptibility-Weighted Images (SWI). The results show that the framework achieves state-of-the-art accuracy, while reducing the annotation time by ∼77% w.r.t. learning-based segmentation methods using pixel-wise labels for training.
Collapse
Affiliation(s)
- Vien Ngoc Dang
- Data Science Department, EURECOM, Sophia Antipolis, France; Artificial Intelligence in Medicine Lab, Facultat de Matemátiques I Informática, Universitat de Barcelona, Spain
| | | | - Rosa Cortese
- Queen Square MS Centre, Department of Neuroinflammation, UCL Queen Square Institute of Neurology, Faculty of Brain Sciences, University College London, UK; Department of Medicine, Surgery and Neuroscience, University of Siena, Italy
| | - Giuseppe Di Giacomo
- Data Science Department, EURECOM, Sophia Antipolis, France; Politecnico di Torino, Turin, Italy
| | - Viola Marconetto
- Data Science Department, EURECOM, Sophia Antipolis, France; Politecnico di Torino, Turin, Italy
| | - Prateek Mathur
- Data Science Department, EURECOM, Sophia Antipolis, France
| | - Karim Lekadir
- Artificial Intelligence in Medicine Lab, Facultat de Matemátiques I Informática, Universitat de Barcelona, Spain
| | - Marco Lorenzi
- Université Côte d'Azur, Inria Sophia Antipolis, Epione Research Group, Valbonne, France
| | - Ferran Prados
- Centre for Medical Image Computing, Department of Medical Physics and Bioengineering, University College London, UK; Queen Square MS Centre, Department of Neuroinflammation, UCL Queen Square Institute of Neurology, Faculty of Brain Sciences, University College London, UK; National Institute for Health Research, University College London Hospitals, Biomedical Research Centre, London, UK; e-health Center, Universitat Oberta de Catalunya, Barcelona, Spain
| | | |
Collapse
|
44
|
Pal A, Xue Z, Desai K, Aina F Banjo A, Adepiti CA, Long LR, Schiffman M, Antani S. Deep multiple-instance learning for abnormal cell detection in cervical histopathology images. Comput Biol Med 2021; 138:104890. [PMID: 34601391 DOI: 10.1016/j.compbiomed.2021.104890] [Citation(s) in RCA: 13] [Impact Index Per Article: 4.3] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/06/2021] [Revised: 09/15/2021] [Accepted: 09/22/2021] [Indexed: 01/18/2023]
Abstract
Cervical cancer is a disease of significant concern affecting women's health worldwide. Early detection of and treatment at the precancerous stage can help reduce mortality. High-grade cervical abnormalities and precancer are confirmed using microscopic analysis of cervical histopathology. However, manual analysis of cervical biopsy slides is time-consuming, needs expert pathologists, and suffers from reader variability errors. Prior work in the literature has suggested using automated image analysis algorithms for analyzing cervical histopathology images captured with the whole slide digital scanners (e.g., Aperio, Hamamatsu, etc.). However, whole-slide digital tissue scanners with good optical magnification and acceptable imaging quality are cost-prohibitive and difficult to acquire in low and middle-resource regions. Hence, the development of low-cost imaging systems and automated image analysis algorithms are of critical importance. Motivated by this, we conduct an experimental study to assess the feasibility of developing a low-cost diagnostic system with the H&E stained cervical tissue image analysis algorithm. In our imaging system, the image acquisition is performed by a smartphone affixing it on the top of a commonly available light microscope which magnifies the cervical tissues. The images are not captured in a constant optical magnification, and, unlike whole-slide scanners, our imaging system is unable to record the magnification. The images are mega-pixel images and are labeled based on the presence of abnormal cells. In our dataset, there are total 1331 (train: 846, validation: 116 test: 369) images. We formulate the classification task as a deep multiple instance learning problem and quantitatively evaluate the classification performance of four different types of multiple instance learning algorithms trained with five different architectures designed with varying instance sizes. Finally, we designed a sparse attention-based multiple instance learning framework that can produce a maximum of 84.55% classification accuracy on the test set.
Collapse
Affiliation(s)
- Anabik Pal
- National Library of Medicine, National Institutes of Health, Bethesda, MD, USA.
| | - Zhiyun Xue
- National Library of Medicine, National Institutes of Health, Bethesda, MD, USA
| | - Kanan Desai
- National Cancer Institute, National Institutes of Health, Bethesda, MD, USA
| | | | | | - L Rodney Long
- National Library of Medicine, National Institutes of Health, Bethesda, MD, USA
| | - Mark Schiffman
- National Cancer Institute, National Institutes of Health, Bethesda, MD, USA
| | - Sameer Antani
- National Library of Medicine, National Institutes of Health, Bethesda, MD, USA
| |
Collapse
|
45
|
Transfer Learning Approach for Classification of Histopathology Whole Slide Images. SENSORS 2021; 21:s21165361. [PMID: 34450802 PMCID: PMC8401188 DOI: 10.3390/s21165361] [Citation(s) in RCA: 10] [Impact Index Per Article: 3.3] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Subscribe] [Scholar Register] [Received: 06/25/2021] [Revised: 08/06/2021] [Accepted: 08/07/2021] [Indexed: 02/07/2023]
Abstract
The classification of whole slide images (WSIs) provides physicians with an accurate analysis of diseases and also helps them to treat patients effectively. The classification can be linked to further detailed analysis and diagnosis. Deep learning (DL) has made significant advances in the medical industry, including the use of magnetic resonance imaging (MRI) scans, computerized tomography (CT) scans, and electrocardiograms (ECGs) to detect life-threatening diseases, including heart disease, cancer, and brain tumors. However, more advancement in the field of pathology is needed, but the main hurdle causing the slow progress is the shortage of large-labeled datasets of histopathology images to train the models. The Kimia Path24 dataset was particularly created for the classification and retrieval of histopathology images. It contains 23,916 histopathology patches with 24 tissue texture classes. A transfer learning-based framework is proposed and evaluated on two famous DL models, Inception-V3 and VGG-16. To improve the productivity of Inception-V3 and VGG-16, we used their pre-trained weights and concatenated these with an image vector, which is used as input for the training of the same architecture. Experiments show that the proposed innovation improves the accuracy of both famous models. The patch-to-scan accuracy of VGG-16 is improved from 0.65 to 0.77, and for the Inception-V3, it is improved from 0.74 to 0.79.
Collapse
|
46
|
Zhang J, Hua Z, Yan K, Tian K, Yao J, Liu E, Liu M, Han X. Joint fully convolutional and graph convolutional networks for weakly-supervised segmentation of pathology images. Med Image Anal 2021; 73:102183. [PMID: 34340108 DOI: 10.1016/j.media.2021.102183] [Citation(s) in RCA: 18] [Impact Index Per Article: 6.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 06/10/2020] [Revised: 07/14/2021] [Accepted: 07/16/2021] [Indexed: 01/18/2023]
Abstract
Tissue/region segmentation of pathology images is essential for quantitative analysis in digital pathology. Previous studies usually require full supervision (e.g., pixel-level annotation) which is challenging to acquire. In this paper, we propose a weakly-supervised model using joint Fully convolutional and Graph convolutional Networks (FGNet) for automated segmentation of pathology images. Instead of using pixel-wise annotations as supervision, we employ an image-level label (i.e., foreground proportion) as weakly-supervised information for training a unified convolutional model. Our FGNet consists of a feature extraction module (with a fully convolutional network) and a classification module (with a graph convolutional network). These two modules are connected via a dynamic superpixel operation, making the joint training possible. To achieve robust segmentation performance, we propose to use mutable numbers of superpixels for both training and inference. Besides, to achieve strict supervision, we employ an uncertainty range constraint in FGNet to reduce the negative effect of inaccurate image-level annotations. Compared with fully-supervised methods, the proposed FGNet achieves competitive segmentation results on three pathology image datasets (i.e., HER2, KI67, and H&E) for cancer region segmentation, suggesting the effectiveness of our method. The code is made publicly available at https://github.com/zhangjun001/FGNet.
Collapse
Affiliation(s)
- Jun Zhang
- Tencent AI Lab, Shenzhen, Guangdong 518057, China
| | - Zhiyuan Hua
- Perception and Robotics Group, University of Maryland, College Park, MD 20742, USA
| | - Kezhou Yan
- Tencent AI Lab, Shenzhen, Guangdong 518057, China
| | - Kuan Tian
- Tencent AI Lab, Shenzhen, Guangdong 518057, China
| | - Jianhua Yao
- Tencent AI Lab, Shenzhen, Guangdong 518057, China
| | - Eryun Liu
- Zhejiang University, Hangzhou, Zhejiang 310027, China
| | - Mingxia Liu
- Department of Radiology and BRIC, University of North Carolina, Chapel Hill, NC 27599, USA
| | - Xiao Han
- Tencent AI Lab, Shenzhen, Guangdong 518057, China.
| |
Collapse
|
47
|
Bi Q, Qin K, Zhang H, Xia GS. Local Semantic Enhanced ConvNet for Aerial Scene Recognition. IEEE TRANSACTIONS ON IMAGE PROCESSING : A PUBLICATION OF THE IEEE SIGNAL PROCESSING SOCIETY 2021; 30:6498-6511. [PMID: 34236963 DOI: 10.1109/tip.2021.3092816] [Citation(s) in RCA: 5] [Impact Index Per Article: 1.7] [Reference Citation Analysis] [Abstract] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 06/13/2023]
Abstract
Aerial scene recognition is challenging due to the complicated object distribution and spatial arrangement in a large-scale aerial image. Recent studies attempt to explore the local semantic representation capability of deep learning models, but how to exactly perceive the key local regions remains to be handled. In this paper, we present a local semantic enhanced ConvNet (LSE-Net) for aerial scene recognition, which mimics the human visual perception of key local regions in aerial scenes, in the hope of building a discriminative local semantic representation. Our LSE-Net consists of a context enhanced convolutional feature extractor, a local semantic perception module and a classification layer. Firstly, we design a multi-scale dilated convolution operators to fuse multi-level and multi-scale convolutional features in a trainable manner in order to fully receive the local feature responses in an aerial scene. Then, these features are fed into our two-branch local semantic perception module. In this module, we design a context-aware class peak response (CACPR) measurement to precisely depict the visual impulse of key local regions and the corresponding context information. Also, a spatial attention weight matrix is extracted to describe the importance of each key local region for the aerial scene. Finally, the refined class confidence maps are fed into the classification layer. Exhaustive experiments on three aerial scene classification benchmarks indicate that our LSE-Net achieves the state-of-the-art performance, which validates the effectiveness of our local semantic perception module and CACPR measurement.
Collapse
|
48
|
Callahan A, Polony V, Posada JD, Banda JM, Gombar S, Shah NH. ACE: the Advanced Cohort Engine for searching longitudinal patient records. J Am Med Inform Assoc 2021; 28:1468-1479. [PMID: 33712854 PMCID: PMC8279796 DOI: 10.1093/jamia/ocab027] [Citation(s) in RCA: 4] [Impact Index Per Article: 1.3] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 10/08/2020] [Accepted: 02/23/2021] [Indexed: 01/02/2023] Open
Abstract
OBJECTIVE To propose a paradigm for a scalable time-aware clinical data search, and to describe the design, implementation and use of a search engine realizing this paradigm. MATERIALS AND METHODS The Advanced Cohort Engine (ACE) uses a temporal query language and in-memory datastore of patient objects to provide a fast, scalable, and expressive time-aware search. ACE accepts data in the Observational Medicine Outcomes Partnership Common Data Model, and is configurable to balance performance with compute cost. ACE's temporal query language supports automatic query expansion using clinical knowledge graphs. The ACE API can be used with R, Python, Java, HTTP, and a Web UI. RESULTS ACE offers an expressive query language for complex temporal search across many clinical data types with multiple output options. ACE enables electronic phenotyping and cohort-building with subsecond response times in searching the data of millions of patients for a variety of use cases. DISCUSSION ACE enables fast, time-aware search using a patient object-centric datastore, thereby overcoming many technical and design shortcomings of relational algebra-based querying. Integrating electronic phenotype development with cohort-building enables a variety of high-value uses for a learning health system. Tradeoffs include the need to learn a new query language and the technical setup burden. CONCLUSION ACE is a tool that combines a unique query language for time-aware search of longitudinal patient records with a patient object datastore for rapid electronic phenotyping, cohort extraction, and exploratory data analyses.
Collapse
Affiliation(s)
- Alison Callahan
- Center for Biomedical Informatics Research, School of Medicine, School of Medicine, Stanford University, Stanford, California, USA
| | - Vladimir Polony
- Center for Biomedical Informatics Research, School of Medicine, School of Medicine, Stanford University, Stanford, California, USA
| | - José D Posada
- Center for Biomedical Informatics Research, School of Medicine, School of Medicine, Stanford University, Stanford, California, USA
| | - Juan M Banda
- Department of Computer Science, Georgia State University, Atlanta, Georgia, USA
| | - Saurabh Gombar
- Department of Pathology, School of Medicine, Stanford University, Stanford, California, USA
| | - Nigam H Shah
- Center for Biomedical Informatics Research, School of Medicine, School of Medicine, Stanford University, Stanford, California, USA
| |
Collapse
|
49
|
Bateson M, Dolz J, Kervadec H, Lombaert H, Ayed IB. Constrained Domain Adaptation for Image Segmentation. IEEE TRANSACTIONS ON MEDICAL IMAGING 2021; 40:1875-1887. [PMID: 33750688 DOI: 10.1109/tmi.2021.3067688] [Citation(s) in RCA: 8] [Impact Index Per Article: 2.7] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 06/12/2023]
Abstract
Domain Adaption tasks have recently attracted substantial attention in computer vision as they improve the transferability of deep network models from a source to a target domain with different characteristics. A large body of state-of-the-art domain-adaptation methods was developed for image classification purposes, which may be inadequate for segmentation tasks. We propose to adapt segmentation networks with a constrained formulation, which embeds domain-invariant prior knowledge about the segmentation regions. Such knowledge may take the form of anatomical information, for instance, structure size or shape, which can be known a priori or learned from the source samples via an auxiliary task. Our general formulation imposes inequality constraints on the network predictions of unlabeled or weakly labeled target samples, thereby matching implicitly the prediction statistics of the target and source domains, with permitted uncertainty of prior knowledge. Furthermore, our inequality constraints easily integrate weak annotations of the target data, such as image-level tags. We address the ensuing constrained optimization problem with differentiable penalties, fully suited for conventional stochastic gradient descent approaches. Unlike common two-step adversarial training, our formulation is based on a single segmentation network, which simplifies adaptation, while improving training quality. Comparison with state-of-the-art adaptation methods reveals considerably better performance of our model on two challenging tasks. Particularly, it consistently yields a performance gain of 1-4% Dice across architectures and datasets. Our results also show robustness to imprecision in the prior knowledge. The versatility of our novel approach can be readily used in various segmentation problems, with code available publicly.
Collapse
|
50
|
Tong Y, Sun Y, Zhou P, Shen Y, Jiang H, Sha X, Chang S. Locating abnormal heartbeats in ECG segments based on deep weakly supervised learning. Biomed Signal Process Control 2021. [DOI: 10.1016/j.bspc.2021.102674] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.3] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 10/21/2022]
|