# Deep Transfer Learning For Whole-Brain fMRI Analyses

## TL;DR

This paper demonstrates that transfer learning significantly improves the decoding of cognitive states from whole-brain fMRI data, especially in small sample clinical settings, by leveraging models pre-trained on large datasets.

## Contribution

It shows that transfer learning from large fMRI datasets enhances model performance on new tasks with limited data, outperforming models trained from scratch.

## Key findings

- Pre-trained models outperform scratch models on new fMRI tasks.
- A pre-trained model correctly decodes 67.51% of cognitive states with only three subjects.
- Transfer learning reduces data requirements for effective fMRI analysis.

## Abstract

The application of deep learning (DL) models to the decoding of cognitive states from whole-brain functional Magnetic Resonance Imaging (fMRI) data is often hindered by the small sample size and high dimensionality of these datasets. Especially, in clinical settings, where patient data are scarce. In this work, we demonstrate that transfer learning represents a solution to this problem. Particularly, we show that a DL model, which has been previously trained on a large openly available fMRI dataset of the Human Connectome Project, outperforms a model variant with the same architecture, but which is trained from scratch, when both are applied to the data of a new, unrelated fMRI task. Even further, the pre-trained DL model variant is already able to correctly decode 67.51% of the cognitive states from a test dataset with 100 individuals, when fine-tuned on a dataset of the size of only three subjects.