Large Connectome Model: An fMRI Foundation Model of Brain Connectomes Empowered by Brain-Environment Interaction in Multitask Learning Landscape
This work addresses the challenge of improving AI model reliability for clinical neuroimaging applications, though it appears incremental by building on existing pretraining methods with multitask learning.
The paper tackled the problem of suboptimal performance in AI models for clinical neuroimaging due to limited sample sizes by developing a foundation model that uses multitask learning with brain-environment interactions, achieving promising results in sex prediction, behavior recognition, and early diagnosis of diseases like Autism and Alzheimer's.
A reliable foundation model of functional neuroimages is critical to promote clinical applications where the performance of current AI models is significantly impeded by a limited sample size. To that end, tremendous efforts have been made to pretraining large models on extensive unlabeled fMRI data using scalable self-supervised learning. Since self-supervision is not necessarily aligned with the brain-to-outcome relationship, most foundation models are suboptimal to the downstream task, such as predicting disease outcomes. By capitalizing on rich environmental variables and demographic data along with an unprecedented amount of functional neuroimages, we form the brain modeling as a multitask learning and present a scalable model architecture for (i) multitask pretraining by tokenizing multiple brain-environment interactions (BEI) and (ii) semi-supervised finetuning by assigning pseudo-labels of pretrained BEI. We have evaluated our foundation model on a variety of applications, including sex prediction, human behavior recognition, and disease early diagnosis of Autism, Parkinson's disease, Alzheimer's disease, and {Schizophrenia}, where promising results indicate the great potential to facilitate current neuroimaging applications in clinical routines.