Rxivist logo

Enhancing multi-center generalization of machine learning-based depression diagnosis from resting-state fMRI

By Takashi Nakano, Masahiro Takamura, Naho Ichikawa, Go Okada, Yasumasa Okamoto, Makiko Yamada, Tetsuya Suhara, Shigeto Yamawaki, Junichiro Yoshimoto

Posted 25 Aug 2019
medRxiv DOI: 10.1101/19004051

Resting-state fMRI has the potential to find abnormal behavior in brain activity and to diagnose patients with depression. However, resting-state fMRI has a bias depending on the scanner site, which makes it difficult to diagnose depression at a new site. In this paper, we propose methods to improve the performance of the diagnosis of major depressive disorder (MDD) at an independent site by reducing the site bias effects using regression. For this, we used a subgroup of healthy subjects of the independent site to regress out site bias. We further improved the classification performance of patients with depression by focusing on melancholic depressive disorder. Our proposed methods would be useful to apply depression classifiers to subjects at completely new sites.

Download data

  • Downloaded 322 times
  • Download rankings, all-time:
    • Site-wide: 102,286
    • In psychiatry and clinical psychology: 433
  • Year to date:
    • Site-wide: 109,072
  • Since beginning of last month:
    • Site-wide: 102,596

Altmetric data

Downloads over time

Distribution of downloads per paper, site-wide