Loading...

GKD: Semi-supervised graph knowledge distillation for graph-independent inference

Ghorbani, M ; Sharif University of Technology | 2021

239 Viewed
  1. Type of Document: Article
  2. DOI: 10.1007/978-3-030-87240-3_68
  3. Publisher: Springer Science and Business Media Deutschland GmbH , 2021
  4. Abstract:
  5. The increased amount of multi-modal medical data has opened the opportunities to simultaneously process various modalities such as imaging and non-imaging data to gain a comprehensive insight into the disease prediction domain. Recent studies using Graph Convolutional Networks (GCNs) provide novel semi-supervised approaches for integrating heterogeneous modalities while investigating the patients’ associations for disease prediction. However, when the meta-data used for graph construction is not available at inference time (e.g., coming from a distinct population), the conventional methods exhibit poor performance. To address this issue, we propose a novel semi-supervised approach named GKD based on the knowledge distillation. We train a teacher component that employs the label-propagation algorithm besides a deep neural network to benefit from the graph and non-graph modalities only in the training phase. The teacher component embeds all the available information into the soft pseudo-labels. The soft pseudo-labels are then used to train a deep student network for disease prediction of unseen test data for which the graph modality is unavailable. We perform our experiments on two public datasets for diagnosing Autism spectrum disorder, and Alzheimer’s disease, along with a thorough analysis on synthetic multi-modal datasets. According to these experiments, GKD outperforms the previous graph-based deep learning methods in terms of accuracy, AUC, and Macro F1. © 2021, Springer Nature Switzerland AG
  6. Keywords:
  7. Deep neural networks ; Diagnosis ; Distillation ; Forecasting ; Graphic methods ; Medical imaging ; Population statistics ; Graph neural networks ; Knowledge distillation ; Medical data ; Multi-modal ; Non-imaging ; Population-based disease prediction ; Semi-supervised ; Semi-supervised graph neural network ; Semi-supervised graphs ; Teachers' ; Graph neural networks
  8. Source: 24th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2021, 27 September 2021 through 1 October 2021 ; Volume 12905 LNCS , 2021 , Pages 709-718 ; 03029743 (ISSN) ; 9783030872397 (ISBN)
  9. URL: https://www.springerprofessional.de/en/gkd-semi-supervised-graph-knowledge-distillation-for-graph-indep/19692410