Computational augmentation of neoplastic endometrial glands in digital pathology displays

Abstract

The pathologic diagnosis of neoplasia requires localization and classification of lesional tissue, a process that depends on the recognition of an abnormal spatial distribution of neoplastic elements relative to admixed normal background tissue. In endometrial intraepithelial neoplasia (EIN), a pre‐cancer usually managed by hysterectomy, a clonally mutated proliferation of cytologically altered glands (‘neoplastic‐EIN’) aggregates in clusters that also contain background non‐neoplastic glands (‘background‐NL’). Here, we used image analysis to classify individual glands within endometrial tissue fragments as neoplastic‐EIN or background‐NL, and we used the distribution of predictions to localize foci diagnostic of EIN. Nuclear coordinates were automatically assigned and were used as vertices to generate Delaunay triangulations for each gland. Graph statistical variables were used to develop random forest algorithms to classify glands as neoplastic‐EIN or background‐NL. Individual glands in an independent validation set were scored by a ‘ground truth’ biomarker (PAX2 immunohistochemistry). We found that exclusion of small glands led to improvement in classification accuracy. Using an inclusion threshold of 200 nuclei per gland, our final model classification accuracy was 77.5% in the validation set, with a positive predictive value of 0.81. We leveraged this high positive predictive value in a point cloud overlay display to assist end‐user identification of EIN foci. This study demonstrates that graph theory approaches applied to small‐scale anatomic elements in the endometrium allow biologic classification by machine learning, and that spatial superimposition over large‐scale tissue expanses can have practical diagnostic utility. We expect this augmented diagnostic approach to be generalizable to commonly encountered problems in other organ systems. © 2020 The Pathological Society of Great Britain and Ireland. Published by John Wiley & Sons, Ltd.

Dynamic changes in AI-based analysis of endometrial cellular composition: Analysis of PCOS and RIF endometrium

“…AI has come to the forefront of digital pathology for pattern recognition and image classification, 27 , 28 which can be utilized to predict the onset and progression of specific diseases in various clinical research fields. 29 , 30 , 31 , 32 , 33 , 34 AI technique is expected to disentangle current impediments in histopathological performance by analyzing massive amounts of whole-slide images within a short amount of time with a high pixel resolution, 35 as demonstrated in this study. Additionally, the AI approach could also assist in bulk RNA sequencing analysis for deconvolution, where adjustment for the epithelial and stromal cellular ratio could be beneficial for identifying cell type composition-dependent and independent gene expression changes in the tissue.…”

AI-algorithm training and validation for endometrial CD138+ cells in infertility-associated conditions; polycystic ovary syndrome (PCOS) and recurrent implantation failure (RIF)

“…In recent years, the advancement of deep CNN technology has revolutionized the rapid analysis of a massive number of WSIs. However, the application of CNN algorithms in uterine studies has predominantly focused on cancer research 18,32 . To our knowledge, our study was the first to target endometrial CD138+ plasma cells in a whole view of slides using a cloud-based CNN platform with high precision and specificity.…”

“…In particular, the artificial intelligence (AI) embodied by CNNs enables rapid analysis of whole tissue slides with high resolution, addressing the current limitations of manual histopathological performance 13 . To date, this promising and potent AI technique has been increasingly applied in various clinical research fields to predict the onset and progression of specific diseases [14][15][16][17][18] . Given that an imbalanced inflammatory milieu has been linked to infertility-associated conditions, such as polycystic ovary syndrome (PCOS) 19 and recurrent implantation failure (RIF) 9,10 , this study utilized an AI algorithm to assess endometrial CD138+ cells in these two conditions.…”

Artificial intelligence deep learning model assessment of leukocyte counts and proliferation in endometrium from women with and without polycystic ovary syndrome

“…To our knowledge, only a few studies have used AI technology in endometrial histologic analysis (21,22). Earlier endometrial studies have been based mostly on randomly chosen images or microscopic fields of samples and manual counting or scoring.…”

“…The objective of the current study was to use an artificial intelligence (AI) deep learning model for the autonomous separation of the endometrial epithelium and stroma, and the subsequent quantification of endometrial leukocytes and proliferation from stained endometrial tissue sections, in an objective and comparable manner. To date, only a few studies have evaluated the performance of AI technology in endometrial analysis although with encouraging results (21,22). From the AI analysis results, the differences in leukocyte counts and proliferation in different cycle phases were evaluated in PCOS and control samples.…”