Multi-pooling 3D Convolutional Neural Network for fMRI Classification of Visual Brain States
Preprint
Abstract
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.Keywords
All Related Versions
This publication has 4 references indexed in Scilit:
- 3D-CNN based discrimination of schizophrenia using resting-state fMRIArtificial Intelligence in Medicine, 2019
- Resting State fMRI Functional Connectivity-Based Classification Using a Convolutional Neural Network ArchitectureFrontiers in Neuroinformatics, 2017
- Automatic Recognition of fMRI-Derived Functional Networks Using 3-D Convolutional Neural NetworksIEEE Transactions on Biomedical Engineering, 2017
- On the Relationship between Classical Grid Search and Probabilistic Roadmaps2004