Pathology Image Analysis Using Segmentation Deep Learning Algorithms

Wang, Shidan; Yang, Donghan M.; Rong, Ruichen; Zhan, Xiaowei; Xiao, Guanghua

doi:10.1016/j.ajpath.2019.05.007

Cited by 261 publications

(148 citation statements)

References 41 publications

Supporting

Mentioning

147

Contrasting

Order By: Relevance

“…The authors' recently published hybrid morphometric and deep learning approach to automating the Paris System utilized a series of specialists semantic segmentation networks for NC ratio calculation requiring thousands of hand annotated images . Improving the quality and automation of cell compartment segmentation could provide significant performance gains to the above and other automated techniques for the analysis of cytological cancer screening tests (Layfield et al, 2017;Wang et al, 2019). Pix2Pix, utilizing a rather small training set, achieved remarkable segmentation performance, yielding an average macro-accuracy of 0.95 and an R 2 value of 0.74±0.019 between the ground truth and predicted NC ratios across the test set (Table 1; Figure 2).…”

Section: Preliminary Resultsmentioning

confidence: 99%

Preliminary Evaluation of the Utility of Deep Generative Histopathology Image Translation at a Mid-Sized NCI Cancer Center

Levy

Jackson

Sriharan

et al. 2020

Preprint

View full text Add to dashboard Cite

Evaluation of a tissue biopsy is often required for the diagnosis and prognostic staging of a disease. Recent efforts have sought to accurately quantitate the distribution of tissue features and morphology in digitized images of histological tissue sections, Whole Slide Images (WSI). Generative modeling techniques present a unique opportunity to produce training data that can both augment these models and translate histologic data across different intra-and-inter-institutional processing procedures, provide cost-effective ways to perform computational chemical stains (synthetic stains) on tissue, and facilitate the creation of diagnostic aid algorithms. A critical evaluation and understanding of these technologies is vital for their incorporation into a clinical workflow. We illustrate several potential use cases of these techniques for the calculation of nuclear to cytoplasm ratio, synthetic SOX10 immunohistochemistry (IHC, sIHC) staining to delineate cell lineage, and the conversion of hematoxylin and eosin (H&E) stain to trichome stain for the staging of liver fibrosis. APPENDIX 1 Mathematical Description of CycleGAN and Pix2Pix Objective FunctionsGANs are tasked with learning a source/target domain Y from latent noise vector Z via mapping G.The loss function for a GAN is specified by:

show abstract

Section: Preliminary Resultsmentioning

confidence: 99%

Preliminary Evaluation of the Utility of Deep Generative Histopathology Image Translation at a Mid-Sized NCI Cancer Center

Levy

Jackson

Sriharan

et al. 2020

Preprint

View full text Add to dashboard Cite

show abstract

“…For example, a GLMNET algorithm 18 would allow the simultaneous selection and estimation of input variable coefficients, at the very least leading to more consistent, if not more accurate, results. More advanced machine learning algorithms, such as neural networks, are now being used in the analysis of images from pathological samples, with new quantification approaches becoming available (reviewed in 10 ). In some cases, deep neural networks have been shown to deliver classifications that are as accurate as those of a specialist, as in the case of skin lesions 11 .…”

Section: Discussionmentioning

confidence: 99%

“…Most papers rely on complex protocols and report only on manually-counted adenoma numbers, or numbers and areas in selected areas of the intestinal tract, although some also include information on adenoma location and size. However, high quality semi-automated methods are now becoming available to facilitate the identification of tumour lesions in histological images 10 , or guide the visual classification of macroscopic tumour lesions including melanomas in patients 11 . Therefore, these methods can offer rapid and objective tumour identification in a broad range of situations.…”

Section: Introductionmentioning

confidence: 99%

A semi-automated technique for adenoma quantification in the Apc^Min mouse using FeatureCounter

Shepherd

Smith

Wakelin

et al. 2019

Preprint

View full text Add to dashboard Cite

Colorectal cancer is a major contributor to death and disease worldwide. The ApcMin mouse is a widely used model of intestinal neoplasia, as it carries a mutation also found in human colorectal cancers. However, the method most commonly used to quantify tumour burden in these mice is manual adenoma counting, which is time consuming and poorly suited to standardization across different laboratories. We describe a method to produce suitable photographs of the small intestine, process them with an ImageJ macro, FeatureCounter, which automatically locates image features potentially corresponding to adenomas, and a machine learning pipeline to identify and quantify them. Compared to a manual method, the specificity (or True Negative Rate, TNR) and sensitivity (or True Positive Rate, TPR) of this method in detecting adenomas are similarly high at about 80% and 87%, respectively. Importantly, total adenoma area measures derived from the automatically-called tumours were just as capable of distinguishing high-burden from low-burden mice as those established manually. Overall, our strategy is quicker, helps control experimenter bias and yields a greater wealth of information about each tumour, thus providing a convenient route to getting consistent and reliable results from a study.

show abstract

“…A number of cell segmentation algorithms have been developed for histopathologic image analysis [17,18]. Several classical ML studies reported that cellular features of H&E staining, nuclear and cytoplasmic texture, nuclear shape (e.g., perimeter, area, tortuosity, and eccentricity) and nuclear/cytoplasmic ratio carry prognostic significance [19][20][21][22].…”

Section: Basics Of Image Analysis: Cellular Analysis and Color Normalmentioning

confidence: 99%

Introduction to digital pathology and computer-aided pathology

Nam

Chong

Jung

et al. 2020

J Pathol Transl Med

View full text Add to dashboard Cite

This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (https://creativecommons.org/licenses/ by-nc/4.0) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited. pISSN 2383-7837 eISSN 2383-7845Digital pathology (DP) is now one of the biggest issues facing the field of pathology. DP is a remarkable innovation that changes the paradigm of microscope-based pathology, which has existed for over 100 years. DP alters the way diagnostic tools represent pathologic images from the microscope to the computer screen and has changed storage media from glass slides to digitalized image files. Digitalized pathologic images stored in computer servers or cloud systems can be transmitted over the Internet, thus changing the temporal and spatial domain of pathologic diagnosis. In addition, machine learning allows software assisting in diagnosis to be developed and applied more actively and effectively.This review describes various concepts related to DP and computer-aided pathologic diagnosis (CAPD), current applications of DP, and various issues related to the implementation of DP. It also briefly introduces the development of computer-aided diagnostic tools and their limitations. DIGITAL PATHOLOGY AND COMPUTER-AIDED PATHOLOGYDP, which initially delineated the process of digitizing whole slide images (WSIs) using advanced slide scanning technology, is now a generic term that includes artificial intelligence (AI)based approaches for detection, segmentation, diagnosis, and analysis of digitalized images [1]. WSI indicates digital representation of an entire histopathologic glass slide at microscopic resolution [2]. Over the last two decades, WSI technology has evolved to encompass relatively high resolution, increased scanner capacity, faster scan speed, smaller image file sizes, and commercialization. The development of appropriate image management systems (IMS) and a seamless interface connection between existing hospital systems such as electronic medical records (EMR), picture archiving communication systems (PACS), and laboratory information systems (LIS) (also referred to as the pathol-Introduction to digital pathology and computer-aided pathology Digital pathology (DP) is no longer an unfamiliar term for pathologists, but it is still difficult for many pathologists to understand the engineering and mathematics concepts involved in DP. Computer-aided pathology (CAP) aids pathologists in diagnosis. However, some consider CAP a threat to the existence of pathologists and are skeptical of its clinical utility. Implementation of DP is very burdensome for pathologists because technical factors, impact on workflow, and information technology infrastructure must be considered. In this paper, various terms related to DP and computer-aided pathologic diagnosis are defined, current applications of DP are discussed, and various issues related to implementation of DP are outlined. The developmen...

show abstract

Pathology Image Analysis Using Segmentation Deep Learning Algorithms

Cited by 261 publications

References 41 publications

Preliminary Evaluation of the Utility of Deep Generative Histopathology Image Translation at a Mid-Sized NCI Cancer Center

Preliminary Evaluation of the Utility of Deep Generative Histopathology Image Translation at a Mid-Sized NCI Cancer Center

A semi-automated technique for adenoma quantification in the Apc^Min mouse using FeatureCounter

Introduction to digital pathology and computer-aided pathology

Contact Info

Product

Resources

About

Pathology Image Analysis Using Segmentation Deep Learning Algorithms

Cited by 261 publications

References 41 publications

Preliminary Evaluation of the Utility of Deep Generative Histopathology Image Translation at a Mid-Sized NCI Cancer Center

Preliminary Evaluation of the Utility of Deep Generative Histopathology Image Translation at a Mid-Sized NCI Cancer Center

A semi-automated technique for adenoma quantification in the ApcMin mouse using FeatureCounter

Introduction to digital pathology and computer-aided pathology

Contact Info

Product

Resources

About

A semi-automated technique for adenoma quantification in the Apc^Min mouse using FeatureCounter