Early Nephrosis Detection Based on Deep Learning with Clinical Time-Series Data.

Journal: Studies in health technology and informatics
Published Date: Aug 21, 2019

Abstract

Nephrosis is disease characterized by abnormal protein loss from impaired kidney. We constructed early prediction model using machine learning from clinical time series data, that can predict onset of nephrosis for more than one month. Long short-term memory capable of recognizing temporal sequential data patterns, was adopted as early prediction model for nephrosis. We verified our proposed prediction model has higher accuracy compared with those of baseline classifiers by 5-fold cross validation.

Authors

  • Yohei Yamasaki
    Graduate School of Informatics Kyoto University, Kyoto-City, Kyoto, Japan.
  • Osamu Sugiyama
    Preemptive Medicine and Lifestyle-related Disease Research Center, Kyoto University Hospital, Kyoto, Japan.
  • Shusuke Hiragi
    Graduate School of Informatics Kyoto University, Kyoto-City, Kyoto, Japan.
  • Shosuke Ohtera
    Graduate School of Informatics Kyoto University, Kyoto-City, Kyoto, Japan.
  • Goshiro Yamamoto
    Kyoto University Hospital, Kyoto-City, Kyoto, Japan.
  • Hiroshi Sasaki
    Kyoto University Hospital, Kyoto-City, Kyoto, Japan.
  • Kazuya Okamoto
    Division of Medical Information Technology and Administration Planning, Kyoto University Hospital, Kyoto, Japan.
  • Masayuki Nambu
    Kyoto University Hospital, Kyoto-City, Kyoto, Japan.
  • Tomohiro Kuroda
    Division of Medical Information Technology and Administration Planning, Kyoto University Hospital, Kyoto, Japan.

Keywords

External Resources

Popular Topics
Recent Journals