Multi-pooling 3D Convolutional Neural Network for fMRI Classification of Visual Brain States
This work addresses the challenge of fMRI-based brain state classification for understanding visual processing mechanisms, representing an incremental improvement in accuracy.
The paper tackled the problem of neural decoding of visual object classification from fMRI data by proposing a multi-pooling 3D convolutional neural network (MP3DCNN), which improved classification accuracy by 1.684% to 14.918% over previous methods for various visual brain state tasks.
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 in decoding brain mechanisms.