In summer 2016, I took part in Project Healthcare, which is a volunteer program at Bellevue Hospital. As volunteers, we completed clinical rotations in Adult, Pediatric, and Psychiatric Emergency Services, Urgent Care, Social Work, the ICU, the Operating Room, and the Cardiac Catheterization Lab.
As a part of this program, I conducted research in the Emergency Department on patients’ tobacco, alcohol, and drug use, and more specifically on a patient’s risk of overdosing from opioids, including heroin, morphine, oxycodone, hydrocodone, etc. We completed a short screener and then gave patients brief advice on how to discontinue their tobacco, alcohol, and/or drug use if they reported risky behavior. If a patient was deemed at risk for an overdose, we dispensed and trained the patient on how to use a medicine called Naloxone (or more commonly known as Narcan), which can help temporarily block the effects of opioids long enough to get the person to a hospital.
In addition, we presented on public health topics at a community health fair. I conducted research on Sexual Assault and Domestic Violence.
Center for Integrative Brain Research
In summer 2015, I had the opportunity to participate in the Neurological Surgery Summer Student Program. I worked with Dr. Nino Ramirez and Dr. Fred Garcia in the Center for Integrative Brain Research (CIBR) at the Seattle Children’s Research Institute (SCRI).
The Ramirez lab studies the PreBötzinger Complex (preBötC), which is an area of the brainstem responsible for generating the respiratory rhythm. In mice, the preBötC can be isolated in a single brain slice and its properties can be studied using electrophysiological techniques. Some of these techniques include intracellular recording, which allows us to record a single neuron from inside a cell; single-electrode extracellular recording, which gives us the population rhythm (essentially an average of all the neurons firing in the preBötC); and multi-electrode recording, which allows us to isolate individual neurons based on their unique waveform in order to observe their network interactions. A few example recordings are shown below. The first picture depicts an intracellular recording (above) and an extracellular recording (below). The second picture displays a multi-electrode recording with several individual units.
The goal of our project was to better understand the preBötC’s response immediately following hypoxia (the augmentation phase). Here is a typical response of the preBötC to hypoxia shown in the extracellular recording below: 1) Eupnea (normal breathing) 2) Augmentation – increase in frequency 3) Steady State – bursting has stopped due to a lack of oxygen 4) Post-Hypoxic Depression 5) Recovery and return to a normal rhythm
Since the augmentation phase has not been previously studied in depth, we quantitatively analyzed this phase (which we defined as the first 200 seconds following hypoxia) and found that, on average, across 10 individual preBötC slices:
- The augmentation phase began at 74% oxygen and ended at 62% oxygen (12% range)
- The Respiratory Period (time between bursts) decreased by 32.4%
- This corresponds to a significant increase in frequency
- The average duration of augmentation was 45 seconds
We also began developing the Slice Vent Technology in order to determine if the preBötC can self-regulate its oxygen levels. The Slice Vent consists of 4 different pumps that circulate artificial cerebral spinal fluid (aCSF) with different concentrations of dissolved oxygen. Two of the pumps are peristaltic pumps that continuously circulate the aCSF in and out of the system. The other two are high-performance liquid chromatography (hplc) pumps: one of the hplc pumps contains a high concentration of oxygen (usually 95%), while the other contains a low concentration of oxygen (either 50% or 0%). Below is a schematic of the Slice Vent System:
The Slice Vent Technology allows isolated neural networks, such as the preBötC, to self-regulate their oxygen environment based on their activity, essentially serving as a set of lungs for the brain slice. As the frequency of the bursting increases, this causes the valves in the system to allow more of the aCSF with a higher concentration of oxygen to enter the slice bath, therefore raising the oxygen tension. The Slice Vent acts as a new tool for understanding the mechanisms most important to neuronal activity during transitions in oxygen tension. Additionally, the Slice Vent helps to explain the neurophysiological hysteresis of the preBötC in response to dynamic fluctuations in oxygen, such as that which we see in many clinical conditions like Sudden Infant Death Syndrome (SIDS).
Over the course of the summer we were able to quantitatively analyze the augmentation phase of the preBötzinger Complex and gain an improved understanding of the mechanisms involved in the neural response to hypoxia. This work will hopefully lead to the development of an improved ventilator, which would use a person’s neural signals to provide a more natural breathing pattern and to promote better health outcomes than the current technology allows for.
Other aspects of the Neurological Surgery Summer Student Program involved OR observations and weekly faculty lectures with presentations from top neuroscience researchers at the University of Washington, Seattle Children’s and Harborview.
DREU Summer Project
In summer 2014, I had the privilege of working with Dr. Helena Mentis at the University of Maryland, Baltimore County through the Distributed Research Experience for Undergraduates (DREU) summer internship. Dr. Mentis is an assistant professor in the Department of Information Systems at UMBC. Here is a link to her website.
I worked on Dr. Mentis’ study involving Parkinson’s patients who have undergone a particular type of surgery called Deep Brain Stimulation (DBS). DBS is a procedure that uses electrodes to target specific areas of the brain, most commonly the subthalamic nucleus (STN), to help regulate many of the worst symptoms of Parkinson’s disease, including tremors, rigidity, bradykinesia (slowness), dyskinesia and gait problems. DBS has been shown to have a significant impact on improving quality of life for many patients and reducing the amount of medication needed on a daily basis. There are different settings on the DBS system that can be customized to each individual patient. These settings include frequency, voltage, pulse width, and the number of leads. Bilateral stimulation, or the use of two leads, is most common for treating Parkinson’s disease, but unilateral stimulation can also be used. Each lead contains four electrodes that can be turned on or off. The electrodes are placed in the desired location in the brain and a small, insulated wire connects the lead to the battery pack implanted in the patient’s chest. The frequency, voltage and pulse width have the ability to affect different Parkinson’s related symptoms and are adjusted as needed during the programming appointments with the clinician. Each patient will respond differently to the various settings and may be more sensitive to a particular setting than another, as no two patients are alike. In addition to the programming sessions, patients are given control to adjust their device outside of the clinic within certain parameters.
As you can imagine, it is very difficult for clinicians to objectively assess symptoms in Parkinson’s patients during these programming appointments. Currently, they must rely on subjective evaluations based on tests that patients complete (such as opening and closing the hand). Frequently, what the clinician observes is not consistent with what the patient experiences. In order to address this problem, we used the Leap Motion sensor to begin replicating some of the tests most commonly used by neurologists to evaluate Parkinson’s symptoms in the clinic. We began with the test of opening and closing the hand.
The Leap Motion SDK has several methods used for skeletal tracking, including a method called grabStrength(), which returns a value in the range of 0 to 1. Zero signifies that the hand is fully open and one indicates that the hand is fully closed. The Leap Motion sensor outputs a value every ten milliseconds, so it is able to very accurately track the hand as it opens and closes. The Leap Motion skeletal tracking is pictured below.
The data gathered from the Leap Motion sensor was output to a CSV file and, from there, we created two different visualizations. The first picture below is a line graph, which shows the grabStrength() value versus time. It is intended to show the “smoothness” of opening and closing the hand, or in other words, how rigid the hand is as it is being opened and closed. The second picture below is a variability graph, which is meant to display the time between each hand opening and closing. It is designed to be an effective way to determine whether the hand is becoming fatigued over time. These visualizations still need to be significantly improved in order to make them useful and understandable for both patients and clinicians.
Over the course of the summer, we were able to develop a digitized hand assessment to evaluate rigidity and slow movement in DBS Parkinson’s patients. The goal with this technology is to aid both patients and clinicians in evaluating symptoms, improving patient engagement in their own treatment, and increasing communication in the clinical setting. Future work will involve improving the visualizations of this assessment so it can be utilized and understood by both the patient and the clinician in order to increase patient involvement in their own care and enhance communication.
Here is a link to a summary of the work. Additionally, here is a link to the paper that I co-authored during my time at UMBC entitled “Getting in Sync: Health and Digital Literacy in Patient Deep Brain Stimulation Device Use”. It was published in the 2014 Workshop on Interactive Systems in Healthcare (WISH) Conference.