False-positive pathology: improving reproducibility with the next generation of pathologists

Mazer, Benjamin L.; Homer, Robert J.; Rimm, David L.

doi:10.1038/s41374-019-0257-2

Cited by 11 publications

(8 citation statements)

References 43 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Similarly, access to an adequate case volume is how we ensure pathology's observational research methods are powered to produce accurate results. 10 Patients, mostly unknowingly, have enabled the education and scholarly work of generations of pathologists by allowing their tissue to remain within institutional archives.…”

Section: Privatized Experiencementioning

confidence: 99%

Protecting the Pathology Commons in the Digital Era

Mazer

Paulson

Sinard

2020

Archives of Pathology &Amp; Laboratory Medicine

Self Cite

View full text Add to dashboard Cite

W hole-slide imaging (WSI) technology and machine learning (ML)-aided diagnostics are primed to bring us new ways of seeing, learning, and understanding human pathology. 1 Physicians, investors, and the public are being promised that these technologies will provide ''faster, more accurate diagnosis of disease'' 2 and allow pathologists to ''make better decisions for their patients.'' 3 We worry, however, that these technological advances also come with practical and ethical risks. One particular concern is that WSI and ML are being accompanied by a monopolization of slides, technology, and intellectual property. Some digital monopolies in this arena are de facto whereas others are legal, but both may lead to a loss of shared access to current and future knowledge. We see 3 long-term risks to the monopolizing forces developing in anatomic pathology: the privatization of seeing, the privatization of experience, and the privatization of understanding. To counter these trends, we recommend that our profession create a digital pathology commons.

show abstract

Section: Privatized Experiencementioning

confidence: 99%

Protecting the Pathology Commons in the Digital Era

Mazer

Paulson

Sinard

2020

Archives of Pathology &Amp; Laboratory Medicine

Self Cite

View full text Add to dashboard Cite

show abstract

“…In these ways, efforts by the cancer research community to characterize mechanisms of metastasis through xenografts, allografts, and genetically engineered animal models are hindered by limited access to pathology expertise, inter-and intra-observer variability, and lack of quantitative and high-throughput methods 22 . Proof-of-principle studies on human tissues have revealed AI-based methods as powerful tools that leverage histological information to classify tumors based on genetic mutations [23][24][25][26] , recurrence, death 27,28 , or response to therapy 29 with high accuracy.…”

Section: Introductionmentioning

confidence: 99%

MetFinder: a neural network-based tool for automated quantitation of metastatic burden in histological sections from animal models

Karz,

Coudray,

Bayraktar

et al. 2023

Preprint

View full text Add to dashboard Cite

Diagnosis of most diseases relies on expert histopathological evaluation of tissue sections by an experienced pathologist. By using standardized staining techniques and an expanding repertoire of markers, a trained eye is able to recognize disease-specific patterns with high accuracy and determine a diagnosis. As efforts to study mechanisms of metastasis and novel therapeutic approaches multiply, researchers need accurate, high-throughput methods to evaluate effects on tumor burden resulting from specific interventions. However, current methods of quantifying tumor burden are low in either resolution or throughput. Artificial neural networks, which can perform in-depth image analyses of tissue sections, provide an opportunity for automated recognition of consistent histopathological patterns. In order to increase the outflow of data collection from preclinical studies, we trained a deep neural network for quantitative analysis of melanoma tumor content on histopathological sections of murine models. This AI-based algorithm, made freely available to academic labs through a web-interface called MetFinder, promises to become an asset for researchers and pathologists interested in accurate, quantitative assessment of metastasis burden.

show abstract

“…In contrast, “ top-down ” approaches involve annotating histopathologic features (histotypes) that have been demonstrated to associate with disease state or outcome (Amin et al, 2017) followed by computation on the multiplexed data to identify underlying molecular patterns. Histopathology has a long history of identifying striking spatial features in small cohorts that do not have prognostic or diagnostic value on follow-up, introducing a note of caution into ‘ bottom-up ’ analysis (Mazer et al, 2019; Voskuil, 2015). At the same time, discoveries arising from ‘t op-down ’ analysis are strongly influenced by prior expectations.…”

Section: Introductionmentioning

confidence: 99%

Multiplexed 3D atlas of state transitions and immune interactions in colorectal cancer

Lin

Wang

Coy

et al. 2021

Preprint

View full text Add to dashboard Cite

Advanced solid cancers are complex assemblies of tumor, immune, and stromal cells that invade adjacent tissue and spread to distant sites. Here we use highly multiplexed tissue imaging, spatial statistics, and machine learning to identify cell types and states underlying morphological features of known diagnostic and prognostic significance in colorectal cancer. We find that a thorough spatial analysis requires imaging the entire tumor region, not small fields of view (e.g. those found in tissue microarrays). When this condition is met, the data reveal frequent transitions between histological archetypes (tumor grades and morphologies) correlated with molecular gradients. At the tumor invasive margin, where tumor, normal, and immune cells compete, localized features in 2D such as tumor buds and mucin pools are seen in 3D to be large connected structures having continuously varying molecular properties. Immunosuppressive cell-cell interactions also exhibit graded variation in type and frequency. Thus, whereas scRNA-Seq emphasizes discrete changes in tumor state, whole-specimen imaging reveals the presence of large- and small-scale spatial gradients analogous to those in developing tissues.

show abstract

False-positive pathology: improving reproducibility with the next generation of pathologists

Cited by 11 publications

References 43 publications

Protecting the Pathology Commons in the Digital Era

Protecting the Pathology Commons in the Digital Era

MetFinder: a neural network-based tool for automated quantitation of metastatic burden in histological sections from animal models

Multiplexed 3D atlas of state transitions and immune interactions in colorectal cancer

Contact Info

Product

Resources

About