Presentation | 2009-06-15 Visual image reconstruction from human brain activity Yoichi MIYAWAKI, Hajime UCHIDA, Okito YAMASHITA, Masa-aki SATO, Yusuke MORITO, Hiroki TANABE, Norihiro SADATO, Yukiyasu KAMITANI, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | Recent functional magnetic resonance imaging (fMRI) studies have shown that visual stimuli that a subject observes can be predicted from fMRI activity patterns using machine learning techniques. However, previous approach has focused on classifying brain activity into pre-specified categories, thereby predicting only simple visual features such as orientation and motion direction, and cannot reconstruct a visual image as it is. Constraint-free visual image reconstruction is more challenging than classification, as it is impractical to specify brain activity for all possible images. Here we reconstructed visual images that subjects observe by combining local image bases of multiple scales, whose contrasts were independently predicted from fMRI activity by automatically selecting relevant voxels and exploiting their correlated patterns. The results suggest that our approach provides an effective means to read out complex perceptual states from brain activity while discovering information representation in multivoxel patterns. We also discuss about future applications of our method to prediction of arbitrary finger/limb movements from human brain activity. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | functional magnetic resonance imaging / vision / reconstruction / machine learning |
Paper # | HIP2009-57 |
Date of Issue |
Conference Information | |
Committee | HIP |
---|---|
Conference Date | 2009/6/8(1days) |
Place (in Japanese) | (See Japanese page) |
Place (in English) | |
Topics (in Japanese) | (See Japanese page) |
Topics (in English) | |
Chair | |
Vice Chair | |
Secretary | |
Assistant |
Paper Information | |
Registration To | Human Information Processing (HIP) |
---|---|
Language | JPN |
Title (in Japanese) | (See Japanese page) |
Sub Title (in Japanese) | (See Japanese page) |
Title (in English) | Visual image reconstruction from human brain activity |
Sub Title (in English) | |
Keyword(1) | functional magnetic resonance imaging |
Keyword(2) | vision |
Keyword(3) | reconstruction |
Keyword(4) | machine learning |
1st Author's Name | Yoichi MIYAWAKI |
1st Author's Affiliation | NICT:ATR Computational Neuroscience Lab.() |
2nd Author's Name | Hajime UCHIDA |
2nd Author's Affiliation | ATR Computational Neuroscience Lab.:Nara Institute of Science and Technology |
3rd Author's Name | Okito YAMASHITA |
3rd Author's Affiliation | ATR Computational Neuroscience Lab. |
4th Author's Name | Masa-aki SATO |
4th Author's Affiliation | ATR Computational Neuroscience Lab. |
5th Author's Name | Yusuke MORITO |
5th Author's Affiliation | National Institute for Physiological Sciences:Sokendai |
6th Author's Name | Hiroki TANABE |
6th Author's Affiliation | National Institute for Physiological Sciences:Sokendai |
7th Author's Name | Norihiro SADATO |
7th Author's Affiliation | National Institute for Physiological Sciences:Sokendai |
8th Author's Name | Yukiyasu KAMITANI |
8th Author's Affiliation | ATR Computational Neuroscience Lab.:Nara Institute of Science and Technology |
Date | 2009-06-15 |
Paper # | HIP2009-57 |
Volume (vol) | vol.109 |
Number (no) | 83 |
Page | pp.pp.- |
#Pages | 6 |
Date of Issue |