Back

Inferring the Joint Distribution of Structural and Functional Connectivity in the Human Brain using UNIT-DDPM

Canamedi, V.

2024-02-06 neuroscience
10.1101/2024.02.03.578750 bioRxiv
Show abstract

AO_SCPLOWBSTRACTC_SCPLOWThe structural wiring of the brain is expected to produce a repertoire of functional networks, across time, context, individuals and vice versa. Therefore, a method to infer the joint distribution of structural and functional connectomes would be of immense value. However, existing approaches only provide deterministic snapshots of the structure-function relationship. Here we use an unpaired image translation method, UNIT-DDPM, that infers a joint distribution of structural and functional connectomes. Our approach allows estimates of variability of function for a given structure and vice versa. Furthermore, we found a significant improvement in prediction accuracy among individual brain networks, implicating a tighter coupling of structure and function than previously understood. Also, our approach has the ad-vantage of not relying on paired samples for training. This novel approach provides a means for identifying regions of consistent structure-function coupling.

Matching journals

The top 2 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.