Technical Basis

Functional ultrasound (fUS) applies high-frequency ultrasound pulses to measure blood flow changes reflecting neural activity, analogous to functional MRI but using ultrasound instead of magnetic resonance. The approach detects microvasculature blood flow changes that correlate with local neural population firing, enabling spatial mapping of brain activity. fUS is a “reading” modality: it captures signals out of the brain without electrical or mechanical stimulation.

Spatial and Temporal Resolution

fUS achieves 2–10 Hz temporal resolution, substantially slower than intracortical recordings (kilohertz) but faster than fMRI (~0.5 Hz). Spatial resolution is approximately 100 micrometers, intermediate between EEG and intracortical approaches. This resolution-speed tradeoff suits many BCI applications while maintaining completely noninvasive recording.

Advantages Over Implants

The fundamental advantage of fUS is that no surgical implantation is required. Ultrasound transducers can be positioned on the scalp similar to EEG electrodes, eliminating infection risks, surgical complications, and chronic inflammatory responses. This noninvasive property could enable widespread BCI adoption if functional decoding capability proves sufficient.

Current Development Status

Merge Labs is the most-capitalised company pursuing fUS as a primary BCI modality, with January 2026 $252M OpenAI-led seed funding and a co-founder roster (Tyson Aflalo, Sumner Norman, Mikhail Shapiro) anchored on foundational fUS BCI research at Caltech. Merge Labs is combining fUS imaging with gene therapy and molecular engineering, in an attempt to read high-bandwidth neural activity without tissue implantation.

Important distinction: fUS vs tFUS

Functional ultrasound (fUS, imaging) is distinct from transcranial focused ultrasound (tFUS, stimulation). tFUS uses focused ultrasound waves to mechanically modulate neural activity at depth, a “writing” modality that delivers stimulation rather than reading signals. Gestala, the Chinese ultrasound BCI startup founded by Tianqiao Chen and Phoenix Peng in 2025/2026, uses phased-array focused ultrasound for therapeutic neuromodulation of deep brain structures (chronic pain, depression, stroke rehabilitation), placing it in the tFUS-stimulation category rather than the fUS-imaging category. Both are noninvasive ultrasound technologies but address opposite ends of the read-write spectrum.

Fundamental Limitations

fUS shares fundamental limitations with other hemodynamic approaches: recording lags neural firing by several seconds due to neurovascular coupling kinetics, and spatial resolution remains coarser than direct electrical recording. Despite limitations, the noninvasive property positions fUS as a promising long-term BCI pathway.