Computational pathology definitions, best practices, and recommendations for regulatory guidance: a white paper from the Digital Pathology Association

The Journal of Pathology, Journal Year: 2019, Volume and Issue: 249(3), P. 286 - 294

Published: July 29, 2019

In this white paper, experts from the Digital Pathology Association (DPA) define terminology and concepts in emerging field of computational pathology, with a focus on its application to histology images analyzed together their associated patient data extract information. This review offers historical perspective describes potential clinical benefits research applications field, as well significant obstacles adoption. Best practices for implementing pathology workflows are presented. These include infrastructure considerations, acquisition training data, quality assessments, regulatory, ethical, cyber-security concerns. Recommendations provided regulators, vendors, practitioners order facilitate progress field. © 2019 The Authors. Journal published by John Wiley & Sons Ltd behalf Pathological Society Great Britain Ireland.

Language: Английский

---

Perspective on Oncogenic Processes at the End of the Beginning of Cancer Genomics

Cell, Journal Year: 2018, Volume and Issue: 173(2), P. 305 - 320.e10

Published: April 1, 2018

Highlights•An overview of PanCancer Atlas analyses on oncogenic molecular processes•Germline genome affects somatic genomic landscape in a pathway-dependent fashion•Genome mutations impact expression, signaling, and multi-omic profiles•Mutation burdens drivers influence immune-cell composition microenvironmentSummaryThe Cancer Genome (TCGA) has catalyzed systematic characterization diverse alterations underlying human cancers. At this historic junction marking the completion over 11,000 tumors from 33 cancer types, we present our current understanding processes governing oncogenesis. We illustrate insights into through synthesis findings TCGA project three facets oncogenesis: (1) driver mutations, germline pathogenic variants, their interactions tumor; (2) tumor epigenome transcriptome proteome; (3) relationship between microenvironment, including implications for drugs targeting events immunotherapies. These results will anchor future rare common primary relapsed tumors, cancers across ancestry groups guide deployment clinical sequencing.Graphical abstract

Language: Английский

---

Deep Learning Models for Histopathological Classification of Gastric and Colonic Epithelial Tumours

Scientific Reports, Journal Year: 2020, Volume and Issue: 10(1)

Published: Jan. 30, 2020

Abstract Histopathological classification of gastric and colonic epithelial tumours is one the routine pathological diagnosis tasks for pathologists. Computational pathology techniques based on Artificial intelligence (AI) would be high benefit in easing ever increasing workloads pathologists, especially regions that have shortages access to services. In this study, we trained convolutional neural networks (CNNs) recurrent (RNNs) biopsy histopathology whole-slide images (WSIs) stomach colon. The models were classify WSI into adenocarcinoma, adenoma, non-neoplastic. We evaluated our three independent test sets each, achieving area under curves (AUCs) up 0.97 0.99 adenocarcinoma respectively, 0.96 adenoma respectively. results demonstrate generalisation ability promising potential deployment a practical histopathological diagnostic workflow system.

Language: Английский

---

Artificial intelligence as the next step towards precision pathology

Journal of Internal Medicine, Journal Year: 2020, Volume and Issue: 288(1), P. 62 - 81

Published: March 3, 2020

Pathology is the cornerstone of cancer care. The need for accuracy in histopathologic diagnosis increasing as personalized therapy requires accurate biomarker assessment. appearance digital image analysis holds promise to improve both volume and precision histomorphological evaluation. Recently, machine learning, particularly deep has enabled rapid advances computational pathology. integration learning into routine care will be a milestone healthcare sector next decade, histopathology right at centre this revolution. Examples potential high-value applications include model-based assessment diagnostic features pathology, ability extract identify novel that provide insights disease. Recent groundbreaking results have demonstrated methods pathology significantly improves metastases detection lymph nodes, Ki67 scoring breast cancer, Gleason grading prostate tumour-infiltrating lymphocyte (TIL) melanoma. Furthermore, models also been able predict status some molecular markers lung, prostate, gastric colorectal based on standard HE slides. Moreover, prognostic (survival outcomes) neural network digitized slides several diseases, including lung melanoma glioma. In review, we aim present summarize latest developments application artificial intelligence

Language: Английский

---

Harnessing multimodal data integration to advance precision oncology

Nature reviews. Cancer, Journal Year: 2021, Volume and Issue: 22(2), P. 114 - 126

Published: Oct. 18, 2021

Language: Английский

---