Development of decoding method to infer visual imagery using ECoG signals

2015 Fiscal Year Final Research Report

• Project/Area Number: 26560467
• Research Category: Grant-in-Aid for Challenging Exploratory Research
• Allocation Type: Multi-year Fund
• Research Field: Brain biometrics
• Research Institution: Osaka University
• Principal Investigator: Takufumi Yanagisawa 大阪大学, 医学(系)研究科(研究院), 助教 (90533802)
• Co-Investigator(Kenkyū-buntansha): KISHIMA Haruhiko 大阪大学, 大学院医学系研究科, 講師 (10332743)
• Co-Investigator(Renkei-kenkyūsha): KAMITANI Yukiyasu 株式会社国際電気通信基礎技術研究所, 脳情報研究所, 室長 (50418513)
• Project Period (FY): 2014-04-01 – 2016-03-31
• Keywords: Decoding / ECoG / MEG / Visual perception / Visual imagery

Outline of Final Research Achievements

The aim of this study is to develop a novel method to infer the semantic contents of the visual perception and visual imagery from the ECoG signals and MEG signals. We recorded both of the ECoG and MEG signals while the subjects were watching the images of 3 or 9 categories. The developed decoding method successfully inferred the categories from the ECoG or MEG signals. Moreover, the same decoding method successfully inferred the semantic contents of the visual imagery which the subject imagined freely. These results are expected to be applied for the severely paralyzed patients to improve their communication ability and the quality of their lives.

Free Research Field

Brain-Machine Interface