34,346 Hits in 5.7 sec

Automatic Feature Selection for Improved Interpretability on Whole Slide Imaging

Antoine Pirovano, Hippolyte Heuberger, Sylvain Berlemont, SaÏd Ladjal, Isabelle Bloch
2021 Machine Learning and Knowledge Extraction  
computing interpretability slide-level heat-maps, based on the extracted features, that improves tile-level classification performances.  ...  Then, after studying the impact of the number of features used for heat-map computation, we propose a corrective approach, relying on activation colocalization of selected features, that improves the performances  ...  Finally, our individual analysis of each feature selected at slide-level enabled us to filter out outlier features, to stabilize interpretability performances, and to automatically select the right number  ... 
doi:10.3390/make3010012 fatcat:7jcmnkajmvc4tmkxfrstk77hte

Imitating Pathologist Based Assessment With Interpretable and Context Based Neural Network Modeling of Histology Images

Arunima Srivastava, Chaitanya Kulkarni, Kun Huang, Anil Parwani, Parag Mallick, Raghu Machiraju
2018 Biomedical Informatics Insights  
the histology whole slide images (WSIs).  ...  Convolutional neural networks (CNNs) have gained steady popularity as a tool to perform automatic classification of whole slide histology images.  ...  activity for each whole slide image.  ... 
doi:10.1177/1178222618807481 pmid:30450002 pmcid:PMC6236488 fatcat:k7fdxa5xfnhapdysv24pmtteqy

Automated Interpretation of Blood Culture Gram Stains by Use of a Deep Convolutional Neural Network

Kenneth P. Smith, Anthony D. Kang, James E. Kirby, Paul Bourbeau
2017 Journal of Clinical Microbiology  
Using an automated microscopy platform, uncoverslipped slides were scanned with a 40× dry objective, generating images of sufficient resolution for interpretation.  ...  After training and validation, we applied the classification algorithm to new images collected from 189 whole slides without human intervention.  ...  In addition, our whole slide 300 scanning protocol was based on selecting pre-defined positions for imaging that were invariant 301 between slides.  ... 
doi:10.1128/jcm.01521-17 pmid:29187563 pmcid:PMC5824030 fatcat:h7ey57wcxvhzxivl3fitxgct2y

Quantitative Spatial Analysis on Whole Slide Images Using U-Net

Sanghoon Lee, Yanjun Zhao, Mohamed Masoud, Saeid Belkasim
2020 Computational Biology and Bioinformatics  
Advances in whole slide imaging technology have promoted a high use of digital slide images and generated a large volume of image data that is reliable and useful in determining treatment outcome.  ...  Features directly learned from raw data are trainable within the deep learning procedure and can be used for the histopathology image classification task.  ...  Results on Spatial Analysis In Section 4.1, we conducted a performance evaluation on tumor and TILs prediction rather than on the whole slide images but on the selected regions.  ... 
doi:10.11648/j.cbb.20200802.18 fatcat:fy73ztj5uncopguo4soutqmemy

Pathology imaging informatics for quantitative analysis of whole-slide images

Sonal Kothari, John H Phan, Todd H Stokes, May D Wang
2013 JAMIA Journal of the American Medical Informatics Association  
Many of these systems and informatics methods still focus on images that represent only limited, manually selected regions of tissue slides rather than on whole-slide images (WSI). 5 By including an  ...  Figure 5 5 Role of region of interest (ROI) selection on the performance of whole-slide image (WSI)-based prediction models. (A) An example WSI.  ... 
doi:10.1136/amiajnl-2012-001540 pmid:23959844 pmcid:PMC3822114 fatcat:uq6g2uymczf4rhujcujltaffxe

An Interpretable Algorithm for Uveal Melanoma Subtyping from Whole Slide Cytology Images [article]

Haomin Chen, T.Y. Alvin Liu, Catalina Gomez, Zelia Correa, Mathias Unberath
2021 arXiv   pre-print
Our method embeds every automatically segmented cell of a candidate cytology image as a point in a 2D manifold defined by many representative slides, which enables reasoning about the cell-level composition  ...  Herein we describe an automated yet interpretable system for uveal melanoma subtyping with digital cytology images from fine needle aspiration biopsies.  ...  An overview of the automatic interpretable algorithm for uveal melanoma subtyping from cytology images.  ... 
arXiv:2108.06246v1 fatcat:hec3dxbwjvg5zobn5pq6jp6j4y

A Visual Latent Semantic Approach for Automatic Analysis and Interpretation of Anaplastic Medulloblastoma Virtual Slides [chapter]

Angel Cruz-Roa, Fabio González, Joseph Galaro, Alexander R. Judkins, David Ellison, Jennifer Baccon, Anant Madabhushi, Eduardo Romero
2012 Lecture Notes in Computer Science  
A method for automatic analysis and interpretation of histopathology images is presented.  ...  The method was evaluated on a challenging problem involving automated discrimination of medulloblastoma tumors based on image derived attributes from whole slide images as anaplastic or non-anaplastic.  ...  from whole slides.  ... 
doi:10.1007/978-3-642-33415-3_20 fatcat:gzsudrkvqfa3vkgwx76oqzsk6i

Localization of Diagnostically Relevant Regions of Interest in Whole Slide Images: a Comparative Study

Ezgi Mercan, Selim Aksoy, Linda G. Shapiro, Donald L. Weaver, Tad T. Brunyé, Joann G. Elmore
2016 Journal of digital imaging  
We use these extracted regions in a visual bag-ofwords model based on color and texture features to predict diagnostically relevant ROIs on whole slide images.  ...  Whole slide digital imaging technology enables researchers to study pathologists' interpretive behavior as they view digital slides and gain new understanding of the diagnostic medical decision-making  ...  ., a member of the Roche Group, for the use of iScan Coreo Au™ whole slide imaging system and HD View SL for the source code used to build our digital viewer.  ... 
doi:10.1007/s10278-016-9873-1 pmid:26961982 pmcid:PMC4942394 fatcat:i7kywtgsofgk3pf4lnchmc2fhm

Automated classification of brain tumor type in whole-slide digital pathology images using local representative tiles

Jocelyn Barker, Assaf Hoogi, Adrien Depeursinge, Daniel L. Rubin
2016 Medical Image Analysis  
An initial surveying stage analyzes the diversity of coarse regions in the whole slide image.  ...  tissue regions in whole slide images that are not directly relevant to the disease may mislead computerized diagnosis algorithms.  ...  No more than one whole slide image came from a single patient. Tiling the images WSI tiling created 1024 × 1024 pixel images at 20× resolution.  ... 
doi:10.1016/ pmid:26854941 fatcat:fy7qn6wpgjbxxjtg3jpu5joxfu

Advances in digital pathology

Evgin Goceri
2018 International Journal of Emerging Trends in Health Sciences  
One of the most successful strategies in pathology is to divide tumors into different subtypes and to adapt the treatment for each tumor.  ...  However, this approach has put a big burden on pathologists, who are reviewing tissue samples under the light of the microscope.  ...  Hand-crafted feature based methods use attributes in images and have a level on image interpretability.  ... 
doi:10.18844/ijeths.v1i2.3107 fatcat:uaqltgtsrjeyvicy522pte2m7e

Effects of annotation granularity in deep learning models for histopathological images [article]

Jiangbo Shi, Zeyu Gao, Haichuan Zhang, Pargorn Puttapirat, Chunbao Wang, Xiangrong Zhang, Chen Li
2020 arXiv   pre-print
slide on deep learning models.  ...  On average, precision, recall and F1-score improves by 7.87%, 8.83% and 7.85% respectively.  ...  The emergence of medical imaging equipment has enabled traditional glass slides to be imaged by a scanner, thus a high quality whole slide images(WSIs) can be obtained.  ... 
arXiv:2001.04663v1 fatcat:73wdxjujebantijjjmjiptq4xm

Improving Interpretability for Computer-aided Diagnosis tools on Whole Slide Imaging with Multiple Instance Learning and Gradient-based Explanations [article]

Antoine Pirovano and Hippolyte Heuberger and Sylvain Berlemont and Saïd Ladjal and Isabelle Bloch
2020 arXiv   pre-print
computing interpretability slide-level heat-maps, based on the extracted features, that improves tile-level classification performances by more than 29% for AUC.  ...  In this paper, we address the question of interpretability in the context of whole slide images (WSI) classification.  ...  We finally proposed explainability heat-maps over whole slides taking into account only identified features. This contribution considerably improved tile classification AUC.  ... 
arXiv:2009.14001v1 fatcat:yriowecbive3jdopuaiknhm5dm


Hongming Xu, Tae Hyun Hwang
2018 bioRxiv   pre-print
Computerized whole slide image analysis is important for assisting pathologists in cancer grading and predicting patient clinical outcomes.  ...  However, it is challenging to analyze whole slide image (WSI) at cellular level due to its huge size and nuclear variations.  ...  Although cellular features can be computed from selected image patches, this may bring the sampling bias for the whole slide image representation.  ... 
doi:10.1101/503094 fatcat:p6l3u5whqjdkberv7ydy3zh3i4

Deep Recurrent Attention Models for Histopathological Image Analysis [article]

Alexandre Momeni, Marc Thibault, Olivier Gevaert
2018 bioRxiv   pre-print
Automatic analysis of pathology images could thus have a significant impact on diagnoses, prognoses and treatment decisions for cancer patients.  ...  However, given it remains intractable to process pathology slides in their entirety, CNNs have traditionally performed inference on small individual patches extracted from the image.  ...  Whole Slide Image Pre-processing We perform the pre-processing steps on the highest slide resolution available (40x magnification).  ... 
doi:10.1101/438341 fatcat:6cv24advmje6nlsseugqqzgtja

A Multi-Channel and Multi-Spatial Attention Convolutional Neural Network for Prostate Cancer ISUP Grading

Bochen Yang, Zhifeng Xiao
2021 Applied Sciences  
Deep learning (DL) is able to automatically extract features from whole-slide images of prostate biopsies annotated by skilled pathologists and to classify the severity of PCa.  ...  A whole-slide image of biopsies has many irrelevant features that weaken the performance of DL models.  ...  Compared to feature-based models [10, 11] , DL models can automatically extract feature information from histopathology whole-slide images and grade the severity of PCa, without the costly step of feature  ... 
doi:10.3390/app11104321 fatcat:v3avawafhrdxxghgxj3gdeiqla
« Previous Showing results 1 — 15 out of 34,346 results