ConvPath: A software tool for lung adenocarcinoma digital pathological image analysis aided by a convolutional neural network

Shidan Wang; Tao Wang; Lin Yang; Donghan M Yang; Junya Fujimoto; Faliu Yi; Xin Luo; Yikun Yang; Bo Yao; ShinYi Lin; Cesar Moran; Neda Kalhor; Annikka Weissferdt; John Minna; Yang Xie; Ignacio I Wistuba; Yousheng Mao; Guanghua Xiao

doi:10.1016/j.ebiom.2019.10.033

ConvPath: A software tool for lung adenocarcinoma digital pathological image analysis aided by a convolutional neural network

EBioMedicine. 2019 Dec:50:103-110. doi: 10.1016/j.ebiom.2019.10.033. Epub 2019 Nov 22.

Authors

Shidan Wang¹, Tao Wang², Lin Yang³, Donghan M Yang¹, Junya Fujimoto⁴, Faliu Yi¹, Xin Luo¹, Yikun Yang⁵, Bo Yao¹, ShinYi Lin¹, Cesar Moran⁶, Neda Kalhor⁶, Annikka Weissferdt⁶, John Minna⁷, Yang Xie⁸, Ignacio I Wistuba⁴, Yousheng Mao⁵, Guanghua Xiao⁹

Affiliations

¹ Quantitative Biomedical Research Center, Department of Population and Data Sciences, University of Texas Southwestern Medical Center, Dallas, TX.
² Quantitative Biomedical Research Center, Department of Population and Data Sciences, University of Texas Southwestern Medical Center, Dallas, TX; Center for the Genetics of Host Defense, University of Texas Southwestern Medical Center, Dallas, TX.
³ Quantitative Biomedical Research Center, Department of Population and Data Sciences, University of Texas Southwestern Medical Center, Dallas, TX; Department of Pathology, National Cancer Center/Cancer Hospital, Chinese Academy of Medical Sciences (CHCAMS), China.
⁴ Department of Translational Molecular Pathology, University of Texas MD Anderson Cancer Center, Houston, TX.
⁵ Department of Thoracic Surgery, National Cancer Center/Cancer Hospital, Chinese Academy of Medical Sciences (CHCAMS), China.
⁶ Department of Pathology, Division of Pathology/Lab Medicine, The University of Texas MD Anderson Cancer Center, Houston, TX.
⁷ Hamon Center for Therapeutic Oncology Research, Department of Internal Medicine and Department of Pharmacology, University of Texas Southwestern Medical Center, Dallas, TX; Harold C. Simmons Comprehensive Cancer Center, University of Texas Southwestern Medical Center, Dallas, TX.
⁸ Quantitative Biomedical Research Center, Department of Population and Data Sciences, University of Texas Southwestern Medical Center, Dallas, TX; Department of Bioinformatics, University of Texas Southwestern Medical Center, Dallas, TX; Harold C. Simmons Comprehensive Cancer Center, University of Texas Southwestern Medical Center, Dallas, TX.
⁹ Quantitative Biomedical Research Center, Department of Population and Data Sciences, University of Texas Southwestern Medical Center, Dallas, TX; Department of Bioinformatics, University of Texas Southwestern Medical Center, Dallas, TX; Harold C. Simmons Comprehensive Cancer Center, University of Texas Southwestern Medical Center, Dallas, TX. Electronic address: Guanghua.Xiao@UTSouthwestern.edu.

Abstract

Background: The spatial distributions of different types of cells could reveal a cancer cell's growth pattern, its relationships with the tumor microenvironment and the immune response of the body, all of which represent key "hallmarks of cancer". However, the process by which pathologists manually recognize and localize all the cells in pathology slides is extremely labor intensive and error prone.

Methods: In this study, we developed an automated cell type classification pipeline, ConvPath, which includes nuclei segmentation, convolutional neural network-based tumor cell, stromal cell, and lymphocyte classification, and extraction of tumor microenvironment-related features for lung cancer pathology images. To facilitate users in leveraging this pipeline for their research, all source scripts for ConvPath software are available at https://qbrc.swmed.edu/projects/cnn/.

Findings: The overall classification accuracy was 92.9% and 90.1% in training and independent testing datasets, respectively. By identifying cells and classifying cell types, this pipeline can convert a pathology image into a "spatial map" of tumor, stromal and lymphocyte cells. From this spatial map, we can extract features that characterize the tumor micro-environment. Based on these features, we developed an image feature-based prognostic model and validated the model in two independent cohorts. The predicted risk group serves as an independent prognostic factor, after adjusting for clinical variables that include age, gender, smoking status, and stage.

Interpretation: The analysis pipeline developed in this study could convert the pathology image into a "spatial map" of tumor cells, stromal cells and lymphocytes. This could greatly facilitate and empower comprehensive analysis of the spatial organization of cells, as well as their roles in tumor progression and metastasis.

Keywords: Cell distribution and interaction; Convolutional neural network; Deep learning; Lung adenocarcinoma; Pathology image; Prognosis.

MeSH terms

Adenocarcinoma of Lung / diagnostic imaging*
Adenocarcinoma of Lung / pathology*
Algorithms
Deep Learning
Female
Histocytochemistry / methods
Humans
Image Processing, Computer-Assisted*
Lung Neoplasms / diagnostic imaging*
Lung Neoplasms / pathology*
Male
Neoplasm Staging
Neural Networks, Computer*
Reproducibility of Results
Software*
Tumor Microenvironment
Web Browser
Workflow

Grants and funding

P50 CA070907/CA/NCI NIH HHS/United States