Back

A non-invasive approach to awake mouse fMRI compatible with multi-modal techniques

Laxer, S.; Eed, A.; Bellyou, M.; Zeman, P.; Gilbert, K. M.; Naderi, M.; Menon, R. S.

2025-01-25 neuroscience
10.1101/2025.01.23.634586 bioRxiv
Show abstract

Mouse functional magnetic resonance imaging (fMRI) studies contribute significantly to basic fundamental and translational neuroscience research. Performing fMRI in awake mice could facilitate complex tasks in the magnet and improve translational validity by avoiding anesthesia-related neural and neurovascular changes. Existing surgical approaches provide excellent motion control but are not desirable for all experiments aiming to scan awake mice. These include studies with transgenic mouse lines that are vulnerable to anesthesia or mice in longitudinal studies involving cognition. To address these needs, we propose a non-invasive restraint to scan mice in the awake state. The restraint was designed to be compatible with brain stimulation and recording approaches often combined with fMRI. It was evaluated on the basis of motion, fMRI data quality, and animal stress levels, and compared to a conventional headpost restraint. We found the proposed approach was effective at restraining mice across a broad range of weights without the need for any anesthesia for setup. The non-invasive restraint led to higher data attrition after censoring high motion volumes, but by acquiring roughly 25% more data we could obtain comparable network spatial specificity to the headpost approach. Our results demonstrate a simple open-source head restraint that can be used for awake mouse fMRI for certain cohorts, and we establish suitable acclimation and scanning protocols for use with this restraint.

Published in Imaging Neuroscience (predicted rank #4) · training set

Matching journals

The top 4 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.