Multi-pooling 3D Convolutional Neural Network for fMRI Classification of Visual Brain States

被引：0

作者：

Zhang, Zhen ^{[1
]}

Takeda, Masaki ^{[2
]}

Iwata, Makoto ^{[1
]}

机构：

[1] Kochi Univ Technol, Sch Informat, Kami, Kochi, Japan

[2] Kochi Univ Technol, Res Ctr Brain Commun, Kami, Kochi, Japan

来源：

2023 IEEE CONFERENCE ON ARTIFICIAL INTELLIGENCE, CAI | 2023年

关键词：

fMRI classification; visual brain states; multipooling 3D convolutional neural network (MP3DCNN);

D O I：

10.1109/CAI54212.2023.00057

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

Neural decoding of visual object classification via functional magnetic resonance imaging (fMRI) data is challenging and is vital to understand underlying brain mechanisms. This paper proposed a multi-pooling 3D convolutional neural network (MP3DCNN) to improve fMRI classification accuracy. MP3DCNN is mainly composed of a three-layer 3DCNN, where the first and second layers of 3D convolutions each have a branch of pooling connection. The results showed that this model can improve the classification accuracy for categorical (face vs. object), face sub-categorical (male face vs. female face), and object sub-categorical (natural object vs. artificial object) classifications from 1.684% to 14.918% over the previous study [1] in decoding brain mechanisms.

引用

页码：118 / 119

页数：2

共 8 条

[1]

Ashburner J., 2014, Functional Imaging Lab-oratory Wellcome Trust Centre for Neuroimaging, P1

[2] On the relationship between classical grid search and probabilistic roadmaps [J].