Predicting Heart Rejection using Histopathological Whole-Slide Imaging and Deep Neural Network with Dropout

Li Tong1, Ryan Hoffman1, Shriprasad Deshpande2, May D. Wang3

  • 1Georgia Institute of Technology
  • 2School of Medicine, Emory University
  • 3Georgia Tech and Emory University

Details

Category

Contributed Paper (Poster)

Theme

5. Imaging Informatics

Sessions

09:05 - 09:55 | Thu 16 Feb | Ballroom D | ThRAF

Rapid Fire Session 01: Imaging Informatics

Abstract

Cardiac allograft rejection is one major limitation for long-term survival for patients with heart transplants. The endomyocardial biopsy is one gold standard to screen heart rejection for patients that have heart transplantation. However, manual identification of heart rejection is expensive and time-consuming. With the development of imaging processing techniques and machine learning tools, automatic prediction of heart rejection using whole-slide images is one promising approach to improve the care of patients with heart transplants. In this paper, we first develop a histopathological whole-slide image processing pipeline to extract features automatically. Then, we construct deep neural networks with and without regularization and dropout to classify the patients into non-rejection and rejection respectively. Our results show that neural networks with regularization and dropout can significantly reduce overfitting and achieve more stable accuracies.

Additional Information

No information added

Video

No videos found