Prediction of heart transplant rejection using histopathological whole-slide imaging

• Type of Material: Article
• Authors: Adrienne E. Dooley, Georgia Institute of Technology; Li Tong, Georgia Institute of Technology; Shriprasad Deshpande, Emory University; Dongmei Wang, Emory University
• Language: English
• Date: 2018-04-06
• Publisher: IEEE Explore
• Publication Version: Author Accepted Manuscript (After Peer Review)
• Copyright Statement: © Copyright 2018 IEEE - All rights reserved.
• Final Published Version (URL): http://dx.doi.org/10.1109/BHI.2018.8333416
• Title of Journal or Parent Work: 2018 IEEE EMBS International Conference on Biomedical and Health Informatics, BHI 2018
• Volume: 2018-January
• Start Page: 251
• End Page: 254
• Grant/Funding Information: This work was supported by the grants from National Institutes of Health (NCI Transformative R01 CA163256, and National Center for Advancing Translational Sciences UL1TR000454), Microsoft Research and Hewlett Packard.; This work was also supported in part by the scholarship from China Scholarship Council (CSC) under the Grant CSC NO. 201406010343.
• Keywords: Histopathology; Surgery; Heart transplantation; Biomedical Engineering; Imaging
• Research Categories: Health Sciences, Radiology; Health Sciences, Medicine and Surgery; Engineering, Biomedical

Abstract

Endomyocardial biopsies are the current gold standard for monitoring heart transplant patients for signs of cardiac allograft rejection. Manually analyzing the acquired tissue samples can be costly, time-consuming, and subjective. Computer-aided diagnosis, using digitized whole-slide images, has been used to classify the presence and grading of diseases such as brain tumors and breast cancer, and we expect it can be used for prediction of cardiac allograft rejection. In this paper, we first create a pipeline to normalize and extract pixel-level and object-level features from histopathological whole-slide images of endomyocardial biopsies. Then, we develop a two-stage classification algorithm, where we first cluster individual tiles and then use the frequency of tiles in each cluster for classification of each whole-slide image. Our results show that the addition of an unsupervised clustering step leads to higher classification accuracy, as well as the importance of object-level features based on the pathophysiology of rejection. Future expansion of this study includes the development of a multi-class classification pipeline for subtypes and grades of cardiac allograft rejection.