Segmentation and Uncertainty Measures of Cardiac Substrates within Optical Coherence Tomography Images via Convolutional Neural Networks

Ziyi Huang, Yu Gan, Theresa Lye, Darnel Theagene, Spandana Chintapalli, Simeran Virdi, Andrew Laine, Elsa D. Angelini, Christine P. HendonISBI 2020 · OpenReview

Abstract

Segmentation of human cardiac tissue has a great potential to provide critical clinical guidance for Radiofrequency Ablation (RFA). Uncertainty in cardiac tissue segmentation is high because of the ambiguity of the subtle boundary and intra-linter-physician variations. In this paper, we proposed a deep learning framework for Optical Coherence Tomography (OCT) cardiac segmentation with uncertainty measurement. Our proposed method employs additional dropout layers to assess the uncertainty of pixel-wise label prediction. In addition, we improve the segmentation performance by using focal loss to put more weights on mis-classified examples. Experimental results show that our method achieves high accuracy on pixel-wise label prediction. The feasibility of our method for uncertainty measurement is also demonstrated with excellent correspondence between uncertain regions within OCT images and heterogeneous regions within corresponding histology images.
BibTeX
@inproceedings{huang2020segmentation,
  author = {Ziyi Huang and Yu Gan and Theresa Lye and Darnel Theagene and Spandana Chintapalli and Simeran Virdi and Andrew Laine and Elsa D. Angelini and Christine P. Hendon},
  title = {{Segmentation and Uncertainty Measures of Cardiac Substrates within Optical Coherence Tomography Images via Convolutional Neural Networks}},
  booktitle = {ISBI},
  year = {2020},
  url = {https://openreview.net/forum?id=_Vt-Qgej4i},
}

Loading related content...

Figure