Semi-Supervised Deep Expectation-Maximization for Low-Dose Pet-Ct

Vatsala Sharma, Ansh Khurana, Sriram Yenamandra, Suyash P. AwateISBI 2022 · OpenReview

Abstract

Reducing the dose of ionizing radiation underlying combined imaging with positron emission tomography (PET) and computed tomography (CT) typically leads to reduced image quality. We propose a novel variational deep-neural-network (DNN) framework for image quality enhancement of low-dose PET-CT images, relying on Monte-Carlo expectation maximization. Unlike existing DNN-based training that pairs low-dose PET-CT images with their corresponding high-dose versions, we propose a semi-supervised learning framework that enables learning using a small number of high-dose images. We propose a robust and uncertainty-aware loss motivated by a heavy-tailed generalized-Gaussian distribution on the residuals between the DNN output and the PET-CT data, aiding our semi-supervised learning scheme. Results on publicly available data show the benefits of our framework, quantitatively and qualitatively, over existing methods.
BibTeX
@inproceedings{sharma2022semi,
  author = {Vatsala Sharma and Ansh Khurana and Sriram Yenamandra and Suyash P. Awate},
  title = {{Semi-Supervised Deep Expectation-Maximization for Low-Dose Pet-Ct}},
  booktitle = {ISBI},
  year = {2022},
  url = {https://openreview.net/forum?id=LHzRoihoWWO},
}

Loading related content...

Figure