Computational discovery of interpretable histopathologic prognostic biomarkers in invasive carcinomas of the breast Restricted; Files & ToC

Abstract

While microscopic examination of tumor resections and biopsies has been a cornerstone in breast cancer grading for decades, it suffers from considerable inter-rater variability due to perceptual limitations and high clinical caseloads. Computational analysis of whole-slide image scans using convolutional neural networks (CNN) can help address this challenge. Unfortunately, CNNs can be difficult to interpret, which motivates our adoption of an approach called concept bottlenecking, where models first detect various tissue structures then use them to make their prediction. Concept bottleneck models require a large set of manual annotation data to train. Unfortunately, manual delineation of histopathologic structures is very demanding and impractical given pathologists’ time constraints. This dissertation describes contributions that fall under the themes of scalable data collection, deep learning-based tissue detection, and the discovery of novel histopathologic biomarkers and associations.

First, we examine crowdsourcing approaches that engage medical students to collect manual annotation data. Our results show that a structured, collaborative approach with pathologist supervision is scalable; the resultant publicly-released BCSS and NuCLS datasets contain 20,000 and 200,000 annotations of tissue regions and nuclei, respectively. We show that medical students produce accurate annotations for predominant, visually distinctive structures and that algorithmic suggestions help scale and improve the accuracy of annotations.

Second, we describe a set of CNN modeling approaches for the accurate delineation of histopathologic structures. We describe various improvements to enhance the performance of nucleus detection CNN models and introduce a technique called Decision Tree Approximation of Learned Embeddings, which helps explain CNN nucleus classifications without compromising prediction accuracy. Additionally, we offer consensus recommendations from the International Immuno-Oncology Working Group surrounding the computational detection of tumor-infiltrating lymphocytes, a critical emerging biomarker. Following these recommendations, we develop and validate a multi-scale CNN model that jointly detects tissue regions and nuclei, employing pre-defined biological constraints to improve accuracy.

Finally, we describe the development of a morphologic signature based on quantitative features extracted from computationally-delineated histopathologic regions and cells. This morphologic signature relies partly on a set of stromal features not captured by clinical guidelines for breast cancer grading, and has a stronger prognostic value.

Table of Contents

This table of contents is under embargo until 11 January 2024

About this Dissertation

Rights statement

• Permission granted by the author to include this thesis or dissertation in this repository. All rights reserved by the author. Please contact the author for information regarding the reproduction and use of this thesis or dissertation.

School	Laney Graduate School
Department	Computer Science and Informatics
Degree	Ph.D.
Submission	Dissertation
Language	English
Research Field	Biology, Bioinformatics Health Sciences, Pathology Computer Science
Keyword	Deep Learning Digital Pathology Big Data Computer Vision Anatomic Pathology Tumor Infiltrating Lymphocytes Computational Pathology Breast Cancer
Committee Chair / Thesis Advisor	Cooper, Lee A. D., Emory University Gutman, David A., Emory University
Committee Members	Gaudet, Mia M., National Institutes of Health Moreno, Carlos S., Emory University

Last modified

Primary PDF

Thumbnail	Title	Date Uploaded	Actions
	File download under embargo until 11 January 2024	2021-12-09 21:32:06 -0500	File download under embargo until 11 January 2024

Supplemental Files

Thumbnail	Title	Date Uploaded	Actions