Research

Mechanistic Study of Transcranial Ultrasound Stimulation (TUS)

There is evidence that ultrasound neuromodulation can become both a new therapeutic tool for the nervous system and a research technique for investigating aspects of brain function. The question of whether ultrasound could be more usefully and safely applied to treat neurological and psychological diseases requires pre-clinical research in laboratory animals, which has been under considerable investigation in our group over the past years. We've demonstrated effective parameters for stimulation in mice, including a frequency dependence. We've shown successful stimulation in several different animal models (monkeys, mice, and sheep), demonstrated how to reduce/eliminate stimulation of the peripheral auditory system, and evaluated the safety histologically in a large animal model. Our current work uses optical calcium imaging in GCaMP modified mice to further these mechanistic studies.

Transcranial Ultrasound Planning

Transcranial ultrasound can be used for stimulation, and also to induce transient or permanent changes in brain tissue by ablation or blood brain barrier opening. For all these applications, as ultrasound propagates through the skull, the bone heterogeneously reflects, refracts, and attenuates the ultrasound waves. This alters the position, intensity, and shape of the beam’s focus. We are working on methods to visualize the focal spot in the brain with MRI. We optimized the method called Acoustic Radiation Force Imaging (MR-ARFI), enhancing the accuracy of the measured displacement and allowing rapid imaging during focused ultrasound therapies. Further, we showed that aberrations that distort the focal spot can be found from the MR-ARFI image and corrected for. In recent papers, we demonstrate that stiffness can be corrected for in MR-ARFI images and that the MR-ARFI displacement can be correlated to the neuromodulatory effect. Going forward, we are working on reducing the pressure of the MR-ARFI method to within the FDA regulatory limit, so that it can be used in human TUS studies.  

Transcranial Ultrasound Simulation

To account for the aberrations caused by the skull (described above) when using a phased array transducer, phase corrections are applied to each transducer element such that the beams constructively interfere at the target of interest. Either you can estimate the acoustic parameters of the skull from CT or MR images (velocity, attenuation) to correct for aberrations (hybrid angular spectrum method, MR vs CT), or you can estimate them from the MR-ARFI image. Moving forward, we are developing a simulation method based on deep learning that renders simulations three orders of magnitude faster than finite difference time domain (FDTD) methods, reducing computation time from tens of minutes to just seconds. This will enable real-time phase aberration corrections and a revisit of the MR-ARFI based method.