BCI image retrieval using fMRI and kernel fuzzy clustering

2014 Fiscal Year Final Research Report

• Project/Area Number: 24500258
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Research Field: Sensitivity informatics/Soft computing
• Research Institution: Kochi University of Technology
• Principal Investigator: YOSHIDA Shinichi 高知工科大学, 工学部, 准教授 (30334519)
• Co-Investigator(Kenkyū-buntansha): HATAKEYAMA Yutaka 高知大学, 医歯学系, 准教授 (00376956) ; OKAMOTO Kazushi 千葉大学, 学内共同利用施設等, 助教 (10615032) ; SAIKI Sachio 神戸大学, その他の研究科, 助教 (40549408)
• Project Period (FY): 2012-04-01 – 2015-03-31
• Keywords: ソフトコンピューティング / BCI / fMRI / 機械学習 / 画像検索 / パターン認識

Outline of Final Research Achievements

This study aims at image retrieval using brain neural data obtained from fMRI---BCI based image retrieval. In order to construct the foundation of BCI image retrieval, we have conducted four brain decoding experiment, color circle, black-and-white shapes, emotional images, and images which have different semantics. We used and compared four machine learning techniques, SVM, backpropagation, random forest, and sparse logistic regression. The result shows that the accuracy of the estimation of image is around 70%.

Free Research Field

ソフトコンピューティング