Statistically unbiased prediction enables accurate denoising of voltage imaging data.

Journal: Nature methods
PMID: 37723246

Abstract

Here we report SUPPORT (statistically unbiased prediction utilizing spatiotemporal information in imaging data), a self-supervised learning method for removing Poisson-Gaussian noise in voltage imaging data. SUPPORT is based on the insight that a pixel value in voltage imaging data is highly dependent on its spatiotemporal neighboring pixels, even when its temporally adjacent frames alone do not provide useful information for statistical prediction. Such dependency is captured and used by a convolutional neural network with a spatiotemporal blind spot to accurately denoise voltage imaging data in which the existence of the action potential in a time frame cannot be inferred by the information in other frames. Through simulations and experiments, we show that SUPPORT enables precise denoising of voltage imaging data and other types of microscopy image while preserving the underlying dynamics within the scene.

Authors

  • Minho Eom
    School of Electrical Engineering, KAIST, Daejeon, Republic of Korea.
  • Seungjae Han
    School of Electrical Engineering, KAIST, Daejeon, Republic of Korea.
  • Pojeong Park
    Department of Chemistry and Chemical Biology, Harvard University, Cambridge, MA, USA.
  • Gyuri Kim
    School of Electrical Engineering, KAIST, Daejeon, Republic of Korea.
  • Eun-Seo Cho
    School of Electrical Engineering, KAIST, Daejeon, Republic of Korea.
  • Jueun Sim
    Department of Materials Science and Engineering, KAIST, Daejeon, Republic of Korea.
  • Kang-Han Lee
    Department of Biology, Chungnam National University, Daejeon, Republic of Korea.
  • Seonghoon Kim
    Department of Biological Sciences, Seoul National University, Seoul 03080, Republic of Korea.
  • He Tian
    Institute of Microelectronics and Tsinghua National Laboratory for Information Science and Technology (TNList), Tsinghua University, Beijing 100084, China.
  • Urs L Böhm
    Department of Chemistry and Chemical Biology, Harvard University, Cambridge, MA, USA.
  • Eric Lowet
    Department of Biomedical Engineering, Boston University, Boston, MA 02215, USA.
  • Hua-An Tseng
    Department of Biomedical Engineering, Boston University, Boston, MA, USA.
  • Jieun Choi
    Graduate School of Medical Science and Engineering, KAIST, Daejeon, Republic of Korea.
  • Stephani Edwina Lucia
    Graduate School of Medical Science and Engineering, KAIST, Daejeon, Republic of Korea.
  • Seung Hyun Ryu
    Interdisciplinary Program in Neuroscience, Seoul National University, Seoul, Republic of Korea.
  • Márton Rózsa
    MTA-SZTE Research Group for Cortical Microcircuits of the Hungarian Academy of Sciences, Department of Physiology, Anatomy and Neuroscience, University of Szeged, Szeged, Hungary.
  • Sunghoe Chang
    Department of Physiology and Biomedical Sciences, Seoul National University College of Medicine, Seoul, South Korea.
  • Pilhan Kim
    Graduate School of Medical Science and Engineering, KAIST, Daejeon, Republic of Korea.
  • Xue Han
    College of Pharmacy, Chengdu University of Traditional Chinese Medicine, Chengdu 611137, China.
  • Kiryl D Piatkevich
    Media Lab, Massachusetts Institute of Technology (MIT), Cambridge, MA, USA.
  • Myunghwan Choi
    Department of Biological Sciences, Seoul National University, Seoul 03080, Republic of Korea.
  • Cheol-Hee Kim
    Department of Biology, Chungnam National University, Daejeon, Republic of Korea.
  • Adam E Cohen
    Department of Chemistry and Chemical Biology, Harvard University, Cambridge, MA, USA.
  • Jae-Byum Chang
    Department of Materials Science and Engineering, KAIST, Daejeon, Republic of Korea.
  • Young-Gyu Yoon
    Media Lab, Massachusetts Institute of Technology (MIT), Cambridge, MA, USA.

Keywords

External Resources

Popular Topics
Recent Journals