A generalizable physiological model for detection of Delayed Cerebral Ischemia using Federated Learning.

Journal: Proceedings. IEEE International Conference on Bioinformatics and Biomedicine
Published Date: Jan 18, 2024

Abstract

Delayed cerebral ischemia (DCI) is a complication seen in patients with subarachnoid hemorrhage stroke. It is a major predictor of poor outcomes and is detected late. Machine learning models are shown to be useful for early detection, however training such models suffers from small sample sizes due to rarity of the condition. Here we propose a Federated Learning approach to train a DCI classifier across three institutions to overcome challenges of sharing data across hospitals. We developed a framework for federated feature selection and built a federated ensemble classifier. We compared the performance of FL model to that obtained by training separate models at each site. FL significantly improved performance at only two sites. We found that this was due to feature distribution differences across sites. FL improves performance in sites with similar feature distributions, however, FL can worsen performance in sites with heterogeneous distributions. The results highlight both the benefit of FL and the need to assess dataset distribution similarity before conducting FL.

Authors

  • Ahmed Elhussein
    Department of Biomedical Informatics, Columbia University, New York Genome Center, New York, NY, USA.
  • Murad Megjhani
    Department of Neurology, Columbia University, New York, NY, USA.
  • Daniel Nametz
    Department of Neurology, Columbia University, New York, NY, USA.
  • Miriam Weiss
    Department of Neurosurgery, RWTH Aachen University, Aachen, Germany.
  • Jude Savarraj
    Department of Neurology, UT Health, Houston, TX, USA.
  • Soon Bin Kwon
    Department of Neurology, Columbia University, New York, NY, USA.
  • David J Roh
    Department of Neurology, Columbia University, New York, NY, USA.
  • Sachin Agarwal
    Department of Neurology, Columbia University, New York, NY, USA.
  • E Sander Connolly
    Department of Neurology, Columbia University, New York, NY, USA.
  • Angela Velazquez
    Department of Neurology, Columbia University, New York, NY, USA.
  • Jan Claassen
    Department of Neurology, Columbia University, New York, NY, USA.
  • Huimahn A Choi
    Department of Neurosurgery, RWTH Aachen University, Aachen, Germany.
  • Gerrit A Schubert
    Department of Neurosurgery, Kantonsspital, Aarau, Switzerland.
  • Soojin Park
    Department of Neurology, Department of Biomedical Informatics, Columbia University, New York, NY, USA.
  • Gamze Gürsoy
    Department of Biomedical Informatics, Department of Computer Science, Columbia University, New York Genome Center, New York, NY, USA.

Keywords

No keywords available for this article.

External Resources

Popular Topics
Recent Journals