Luis Hernandez-Garcia, PhD

 

Research Professor

 

 

Biomedical Engineering

and MRI Research Facility

University of Michigan

 

 

1096 BIRB

2360 Bonisteel St.

  Ann Arbor, MI 48109-2108

 Phone: (734) 763 9254

 Fax:   (734) 936 4218

 

 

My contribution to scientific research is primarily in the study of human brain function, although I have explored other topics in biomedical engineering research as well.  More specifically, I am focused on developing much-needed tools for studying human brain function non-invasively.   The bulk of neuroscience experiments, generally speaking, consist of stimulating the brain in some way and then  observing its response.  This strategy is very challenging when studying the awake human brain for obvious ethical reasons. However, in the last two decades we have witnessed a revolution in the available technologies to stimulate and observe brain responses. Functional brain imaging has been such a major breakthrough in the last twenty or thirty years, because it has made it possible to observe the responses to physical stimuli and cognitive challenges non-invasively.  Numerous MRI compatible audio-visual and tactile technologies now exist to stimulate the brain of research participants while they are being scanned.  Additionally, new non-invasive technologies have been developed to stimulate the brain directly through electrical, magnetic and ultrasound pulses.

 

I have been fortunate enough in my career to be a small part of that revolution.  I have focused on magnetic resonance imaging (MRI) methods for imaging the brain’s activity and its changes in response to stimulus, whether direct stimulation or cognitive challenge.  In the area of direct brain stimulation, I have worked on the development of transcranial magnetic stimulation (TMS) and transcranial ultrasound stimulation (TUS) techniques. More specifically, my publication record and contributions to the field are in the following areas:

 

(1)  Non-invasive study of cerebral hemodynamics. 

 

The vascular response to neuronal activity constitutes a powerful indicator of brain activity.  Thus, it is crucial that we understand the relationship between neuronal activity, metabolism and blood flow.  Additionally, blood flow related parameters are also very informative about tissue health, so blood flow imaging is an extremely powerful diagnostic tool in the clinic. Thus, the main thrust of my research has been dedicated to the development of MRI based measurements of vascular parameters, such as perfusion and blood volume, and to understanding the hemodynamic response to brain activation. In this regard, I have worked on the development of mathematical models of the BOLD response to neuronal stimulation.  I have also developed several variants of the Arterial Spin Labeling (ASL) technique to measure different parameters related to the vascular responses, such as perfusion, transit time and arterial blood volume in an efficient manner.  This work involved MRI acquisition methods development and modeling of the vascular bed and the MRI signal

 

(2) Development of Functional MRI analysis. 

 

I have been very active in developing techniques for the analysis of functional MRI data, particularly those data collected with the ASL methods mentioned above.  In that regard, I have worked on the development a modeling framework to detect and quantify perfusion changes from ASL based FMRI time series data with maximal efficiency.  This work included a statistical framework for the estimation and detection of brain activity from ASL data.  It also included methods to take advantage of the information stored in the phase images in the data, and methods to estimate quantitative vascular parameters related to brain activity from ASL data collected during an activation paradigm. 

 

I have also worked on analysis techniques for more the more “standard” FMRI data, which is based on the Blood Oxygenation Level Dependent (BOLD) effect.  I first focused on understanding the temporal sensitivity of the BOLD effect in order to carry out causal network analysis, but more recently, I have developed a statistical framework for identifying the hierarchy and directionality of the information flow within a brain network by examining statistical measures of sufficiency and necessity.  See for examples peer reviewed pubs.

 

(3) Development of non-invasive, direct, transcranial stimulation.

 

Transcranial magnetic stimulation (TMS) is a powerful tool to stimulate the brain directly and non-invasively by applying electrical field pulses to the brain with a powerful electromagnet.  These electromagnetic field pulses can be used to induce or inhibit neuronal activity, depending on the frequency at which they are applied.  A major challenge facing TMS is that its ability to penetrate the brain is limited by the physics of low frequency electromagnetic fields: they decay rapidly with the distance to the source.  The second limitation is its targeting capability because often one stimulates brain tissue that is not part of the target in order to achieve the penetration depth required to reach the target.  As a result, current designs of TMS coils can only achieve limited penetration depths and are difficult to focus because of the formation of secondary fields.

 

Motivated by these challenges, I have made several contributions in the area of transcranial magnetic stimulation (TMS), primarily by developing new designs and design strategies for TMS coils, as well as methods for targeting and for evaluation of the instrument’s performance. My collaborators and I have been able to design probes that can dramatically improve the depth penetration of TMS probes while dramatically reducing the volume of unwanted tissue stimulation. One of these designs was based on the use of actively shielded probes, and the other one consisted of an array of probes that is driven by a single current source.  Additionally I have also developed an MRI based method to image the magnetic fields generated by TMS coils in order to improve the targeting and dosimetry of TMS.  This allows the user to verify the region of tissue that will be stimulated by a TMS coil.  These techniques can also be applied to radiofrequency ablation therapy.

 

An intriguing alternative to TMS for direct brain stimulation is the use of ultrasonic vibrations through the skull in order to facilitate or inhibit action potentials.  In collaboration with the BME department’s ultrasound group, I am currently working on the theory and methods to combine ultrasound waves and magnetic fields in order to induce or inhibit neuronal excitation through the skull completely non-invasively.  The advantage of this technique over TMS based techniques is that it would allow much deeper penetration of the stimulus into the brain, as well as allow for much more precise targeting of the stimulated tissue.  This is a work in progress.

 

(4) Study of cognitive interventions.  

 

I have employed the technical developments above to study how of cognitive training and other interventions affect brain function and help the brain “re-map” itself.  For example, ASL techniques allowed my collaborators and I to find how working memory training for a two-week period can remap brain activity.  We observed how the distribution of brain activity at baseline and alter the responses during an activation challenge by observing the corresponding blood flow changes.

 

I am currently conducting similar studies on depressed patients in order to identify which networks are affected by TMS therapy.  The study uses the ASL techniques described earlier in order to identify brain changes induced by the treatment.  TMS and ASL are a natural fit because ASL can be used to measure the changes in the brain networks activity levels that occur as a result of TMS treatment, and help us understand how the brain re-maps itself as a response to the treatment.

 

(5) Development of MRI guided histotripsy. 

 

Although unrelated to brain function, I am also involved in several projects aiming to characterize and guide a new ultrasound ablation therapy known as “histotripsy”.  Histotripsy is an ultrasound technology developed at UM which allows to liquefy tissue with ultrasonic vibrations by inducing cavitation events in a focal spot within the tissue.  The technique offers tremendous potential for non-invasive removal of tumors, among other applications.

 

My colleagues and I have developed several MRI based methods to provide targeting feedback and dosimetry for histotripsy treatments.  We began by examining the changes in MR properties of histotripsy treated tissue and developing MRI protocols to identify histotripsy lesions.   We then developed a method to observe the cavitation bubbles that occur during cavitation, in order to use MRI imaging to steer the histotripsy intervention. This work is still an active area of research.