Scalable Computational Pathology: From Interactive to Deep Learning Öffentlichkeit

Abstract

Advances in microscopy imaging and genomics have created an explosion of patient data in the pathology domain. Whole-slide images of histologic sections contain rich information describing the diverse cellular elements of tissue microenvironments. These images capture, in high resolution, the visual cues that have been the basis of pathologic diagnosis for over a century. Each whole-slide image contains billions of pixels and up to a million or more microanatomic objects whose appearances hold important prognostic information. Combining this information with genomic and clinical data provides insight into disease biology and patient outcomes. Yet, due to the size and complexity of the data, the software tools needed to allow scientists and clinicians to extract insight from these resources are non-existent or limited. Additionally, current methods utilizing humans is highly subjective and not repeatable. This work aims to address these shortcomings with a set of open-source computational pathology tools.

We first present a comprehensive interactive machine learning framework for assembling training sets for the classification of histologic objects. The system provides a complete infrastructure capable of managing the terabytes worth of images, object features, annotations and metadata in real-time. Active learning algorithms are employed to allow the user to work in tandem with the system in an intuitive and efficient manner. We demonstrate how the system can be used to phenotype microvascular structures in gliomas to predict survival, and to explore the molecular pathways associated with these phenotypes. Quantitative metrics are developed to describe these structures.

We also present a scalable, high-throughput, deep convolutional learning framework for the classification of histologic objects is presented. Due to its use of representation learning, the framework does not require the images to be segmented, instead learning optimal task-specific features in an unbiased manner. Addressing scalability, the graph-based, parallel architecture of the framework allows for the processing of large image archives consisting of hundreds of slides and hundreds of millions of histologic objects. We demonstrate the system's capabilities classifying cell nuclei in lower grade gliomas.

Table of Contents

1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

1.1 Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

1.2 Contributions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4

1.3 Organization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5

2 Active Learning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6

2.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6

2.2 Active Learning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7

3 HistomicsML - A Scalable Interactive Learning Framework..12

3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12

3.2 Architecture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14

3.2.1 Random Forests . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14

3.2.2 Data Formats . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16

3.2.3 Image Segmentation and Feature Extraction . . . . . . . . . . . . 17

3.2.4 Scalability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18

3.2.5 Learning process . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20

3.2.6 Workflow . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23

3.2.7 Command line tools . . . . . . . . . . . . . . . . . . . . . . . . . . 32

3.3 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32

4 Microvascular proliferation and hypertrophy analysis . . 35

4.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35

4.2 Principal Curves . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36

4.3 Phenotyping Microvascular Structures in Gliomas . . . . . . . . . 38

4.4 Microvascular phenotypes accurately predict survival . . . . . . 42

4.5 Genomic Integration Identifies Phenotype Associated Pathways. . 46

4.6 Active learning training improves prognostication . . . . . . . . . . . . . 48

5 Deep Learning. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50

5.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50

5.2 Articial Neural Networks. . . . . . . . . . . . . . . . . . . . . . . . . . . 52

5.3 Convolutional Neural Networks. . . . . . . . . . . . . . . . . . . . . . . . 53

5.4 Deep Convolutional Neural Networks Augmentation . . . . . . . . . . . . 56

6 TissueNet - A Scalable Deep Learning Framework. . . . . 59

6.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59

6.2 Architecture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61

6.2.1 Nuclear Detection and Image Normalization . . . . . . . . . . . . 61

6.2.2 Software . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61

6.2.3 Networks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64

6.3 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67

7 Discussion and Future Work. . . . . . . . . . . . . . . . . . . . . . 70

7.1 Discussion. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70

7.2 Future work. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72

Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74

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	PhD
Submission	Dissertation
Language	English
Research Field	Computer Science
Stichwort	Active Learning Computational Pathology Biomedical Informatics
Committee Chair / Thesis Advisor	Cooper, Lee, Emory University
Committee Members	Brat, Daniel J, Emory University Sunderam, Vaidy S, Emory University Gutman, David Andrew, Emory University

Zuletzt geändert

Primary PDF

Thumbnail	Title	Date Uploaded	Actions
	Scalable Computational Pathology: From Interactive to Deep Learning ()	2018-08-28 15:44:21 -0400	Press to Select an action Download

Supplemental Files

Thumbnail	Title	Date Uploaded	Actions