Abstract
Functional ultrasound imaging (fUS) is an emerging technique that indirectly measures neural activity via changes in blood volume. To date, it has not been used to image chronically during cognitive tasks in freely moving animals. Performing those experiments faces a number of exceptional challenges: performing large durable craniotomies with chronic implants, designing behavioral experiments matching the hemodynamic timescale, stabilizing the ultrasound probe during freely moving behavior, accurately assessing motion artifacts, and validating that the animal can perform cognitive tasks at high performance while tethered. In this study, we provide validated solutions for those technical challenges. In addition, we present standardized step-by-step reproducible protocols, procedures, and data processing pipelines that open up the opportunity to perform fUS in freely moving rodents performing complex cognitive tasks. Moreover, we present proof-of-concept analysis of brain dynamics during a decision-making task. We obtain stable recordings from which we can robustly decode task variables from fUS data over multiple months. Moreover, we find that brain-wide imaging through hemodynamic response is nonlinearly related to cognitive variables, such as task difficulty, as compared to sensory responses previously explored.
Introduction
Systems neuroscience aims to understand how brain-wide dynamics produce complex behavior. While tools like fMRI offer deep-brain penetration, they typically require animals to be immobilized in high-magnetic-field environments, which limits the study of naturalistic behaviors. Conversely, optical imaging and electrophysiology are often constrained by depth or spatial resolution. Functional ultrasound (fUS) imaging serves as a promising middle ground, measuring cerebral blood volume as a proxy for neural activity. However, using fUS in freely moving animals over long periods has historically been difficult due to the need for large, durable craniotomies, probe stabilization, and robust motion artifact correction.
Methods and Implementation
Surgical Innovation and Probe Stabilization
The researchers developed a "full-stack" solution to the mechanical challenges of chronic imaging. This includes a protocol for a large (8 mm x 19 mm) chronic craniotomy protected by a custom-designed, 3D-printed headplate and probe holder. This setup remains stable for months, allowing for longitudinal studies in the same animal.
Behavioral Paradigm Adaptation
Standard rodent decision-making tasks often occur in under a second, which is too fast for the slow temporal dynamics of hemodynamic responses. To address this, the team modified the "Poisson Clicks" evidence accumulation task. Rats were trained to "fixate" and listen to auditory stimuli for 3–5 seconds, stretching the behavioral timescale to match the fUS imaging capabilities.
Computational Pipeline
The study introduced a specialized data-processing toolkit:
- Motion Correction: A rigid motion model combined with a new metric called MEPS (Motion Effected Pixel Statistics) was used to quantify data distortion.
- GraFT (Graph-Filtered Time-trace): A dictionary-learning approach that decomposes fUS video into independent spatial profiles and time-traces without assuming anatomical boundaries, allowing for functional parcellation.
Results and Validation
Behavioral Performance and Stability
A key finding of the study was that rats could carry a 15g tethered probe and perform complex, high-level cognitive tasks at the same proficiency as untethered animals. Psychometric curves showed no significant performance degradation due to the tethering system.
Long-term Reliability and Decoding
The researchers achieved stable imaging planes over 60+ days, with minimal anterior-posterior shift between sessions months apart. Using the decomposed fUS signals and Support Vector Machines (SVMs), they robustly decoded whether a rat was in a trial or between trials. More impressively, they could decode task difficulty (hard vs. easy trials) with high accuracy. Interestingly, cognitive variables like task difficulty showed a nonlinear relationship with the brain-wide hemodynamic response, with nonlinear SVMs significantly outperforming linear ones.
Discussion and Future Directions
This study provides a reproducible framework for mesoscale brain imaging in behaving rodents. While the current motion correction assumes rigid motion and is limited to 2D slices, it establishes a foundation for longitudinal cognitive research. The authors suggest that the next logical step is combining this global fUS "screening" with targeted, local recordings—such as electrophysiology or optogenetics—to bridge the gap between mesoscale blood-flow dynamics and microscale neuronal firing.
Summary updated