Category: Summer Team Impact Project

Pupil as an Indicator for Neuropathic Pain

Author(s): Justin Matthews, Venkat Kalyan Reddy Yasa

Mentor(s): Nathalia Peixoto, Electrical and Computer Engineering

While brightness, near fixation, and arousal/mental effort are the main three kinds of stimuli to which human pupil (herein after pupil) may react to, people were on a search for more ways that may affect the pupil size. Previous research suggests that pupil diameter increases when exposed to heat stimuli. While that was based on external stimuli, there are chances that similar pattern may occur with internal stimuli (pain from inside). As far as our knowledge extends, there was no previous research conducted on pupil response from internal pain. This paper presents how pupil size increases 22.4 – 33.1% more than normal pupil size under internal stimuli./expand]

Opening
Hello, for the Summer Team Impact Program, Venkat Kalyan Reddy Yasa and I, Justin Matthews, worked on this project titled Pupil as an Indicator for Neuropathic Pain. It is under the scientific field of pupillometry, but what exactly is pupillometry?

First slide
It is the study of changes in the diameter of the pupil as a function of cognitive processing. To put it in simpler terms, it is the study of how the pupil changes size due to different stimuli. Research in this field began as early as the 1960s with the first researchers being Eckhard H. Hess and James M. Polt. Their first paper, published in 1960, found that the pupil size increases when viewing emotional or interesting visual stimuli. As cheaper and more reliable eye tracking equipment became accessible, pupillometry research has had a renaissance within the last two decades. Now that you know some of the history behind pupillometry, let’s look at some of the past research.

Second slide
Researchers have narrowed down pupil response to three distinct kinds of stimuli. The pupil will tend to change size in response to brightness, near fixation, and increases to arousal and mental effort. For brightness, researchers were looking at how the pupil responds to different amounts of light. For near fixation, researchers were looking at how the pupil responds when fixating on an object at different distances. For arousal/mental effort, researchers were looking at how the pupil responds when focusing on a task. While these are the main three stimuli, there is another one that is currently being looked into and that is pain.

Third Slide
Experiments conducted that were observing the relation between pupil dilation and pain concluded that the pupil diameter increases as a response to pain. The goal of our experiment is to support that idea, with the difference being the way the pain stimulus is inflicted on the participants.

Fourth Slide
We had a very simple set up for this experiment. An external webcam was taped to the top of the desktop and the recording was done with the camera app. Participants sit on a chair and position themselves so that their eye is in frame. A one minute and thirty second video is then recorded of the participant’s eye. The first thirty seconds is them without pain, while the remainder of that video is the participant in pain after eating 1.2 grams of Tabasco hot sauce. After separating the sections of the film and cropping them we got the length of the pupil (in pixels) in each frame. We then calculated the median length of the pupil without pain and counted how many times the pupil length (while in pain) went over and under the median.

Fifth Slide
Here is an example of one of the graphs created. The x-axis is the pixel length and the y-axis is the number of times the pupil was at that length. Looking at the “w/o pain” graph, we can see that the median is roughly here. If we put the same median on the “w/ pain” graph, it is clear to see that the pupil has more instances of dilating over the median. This supports the finding in previous research as the pupil size goes over the normal average size repeatedly in response to pain.

Sixth Slide
Thank you for your time and thanks to Mason and the OSCAR office for this opportunity. If you want to know more about our project you can visit the following website.

College of Science Summer Team Impact Project

Finger Detection Movement Project

Author(s): Justin Matthews

Mentor(s): Luciano de Oliveira Neris, Information Technology

Object detection is a form of machine learning that utilizes computer vision and deep learning techniques. It can be used to identify targeted objects in images and videos with precise accuracy. For detecting hand gestures, the method most commonly used is computer vision. When using computer vision, there are some factors to consider that could impact the results of the experiment. Due to this, a variety of different techniques were developed in order to mitigate the significance those factors have on the data. This paper will present one technique to use when creating a machine to detect hand gestures

Hello, my name is Justin Matthews, and for the Summer Team Impact Program I worked on this project titled Finger Detection Movement.

For some background knowledge, Machine learning is “a field of study that gives computers the ability to learn without being explicitly programmed”. The origin of machine learning begins with a psychologist named Frank Rosenblatt from Cornell University. He based the design of the machine on the human nervous system. The machine was named “perceptron” and its purpose was to recognize the letters of the alphabet. A task that falls under machine learning is called object detection. Object detection is a form of machine learning that utilizes computer vision and deep learning techniques. It can be used to identify targeted objects in images and videos with precise accuracy. For my project, I created a machine to detect hand gestures during a real time video streamed from a webcam.

Materials that I used for this project include: Visual Studio Code, Anaconda Prompt, Tensorflow, Python, and OpenCV. Visual Studio Code was used to edit my code. Anaconda Prompt was used to run my code. Tensorflow in combination with Keras, was used to save and load the data of the model that was created and trained. Python was used to code the project. Lastly, OpenCV was used to access my laptop’s webcam and display real time video coming from it.

The next portion of this video will be a demonstration on how my program works. What you are seeing now is Anaconda Prompt. This is the normal set up I do before starting to code, test, etc. What I am doing is going into the proper directory and checking to make sure I am on the right path. You can see that I have a folder named dataset and when I enter and display the contents, notice how it is empty. That is because we have to populate that folder which is what my “builDataset” script does. I have to type python *script name* *name I want the image folder to be* *# of images*. Unfortunately I am unable to record the window that appears but as you can see by the next line, 300 images are saved. Once I open my dataset folder you can see that a new folder title Zero appears and in it are all the photos that were taken. I repeat this process for the rest of the numbers (1-5). Once all of that is done it is now time to build and train the model. I simply run the command that you see there and a bunch of text that you can ignore appears. You can see that a loading bar will appear and next to it loss and accuracy. Accuracy is pretty self explanatory and that just means how accurate your model is at guessing the finger position and loss is how much data was lost during that training period. In total there are 10 total training periods each taking roughly 40 seconds each. But with the power of movie magic we can cut all of that. Once the training is complete a graph of the accuracy and lost will be displayed (but my recording software couldn’t record that). Now we can test to see if the model is accurate. First I’ll be testing if it can detect the number three which is does successfully. Let’s try it again but with a different photo. Again success. Now let’s see if it can detect zero. Success once again

These are the graphs that display after the training is complete. On the left is the accuracy graph and on the right is the loss graph. As the accuracy of the model increases, the data loss during training decreases. This means that accuracy and loss have an inverse relationship with one another.

This next screenshot is the window that appears after running the buildDataset script. The way it works is simple as you hold up your hand so that it is within the parameter of the rectangle. You then press the ‘a’ key to start taking screenshots. To pause the screenshot process you just hit the ‘a’ key again. To exit the window early, just press the ‘q’ button. Once the specified number of screenshots are taken, the window will automatically close.

Thank you for your time and thanks to Mason and the OSCAR office for this opportunity. If you want to know more about my project you can visit the following websites: https://detect.onmason.com/

College of Engineering and Computing Honors College Summer Team Impact Project

Robotic Mirror Therapy Exoskeleton

Author(s): Omar Ahmed Alsuhaibani, Franklin Farrel, German Kuznetsov, Elizabeth Kwari, Justin Matthews, Sara Razavi

Mentor(s): Nathalia Peixoto, Engineering

Hemiplegic cerebral palsy (CP) is a type of unilateral CP that causes paralysis on only one side of an individual’s body. While there is no cure for a disorder such as hemiplegic CP, there are several technological solutions available that can assist these children in maintaining balance and improving muscle movement. However, such methods do come at a cost, one that not all people are able to afford. Our team’s project for the Summer 2022 STIP program is focused on creating a robotic mirror therapy-based exoskeleton for children with hemiplegic cerebral palsy – the second most common form of CP that affects the sensorimotor function of the limbs on one side of the body. The overall design of the exoskeleton includes one arm that is motor-based and the other arm that is sensor-based. Our team chose robotic mirror therapy as the subject of our exoskeleton because it shows promise in assisting individuals with hemiplegic cerebral palsy, helping to improve the motor functions pertaining to the impaired side of their body. For this project, our team utilized cost-effective materials, including 3D printing parts, to create our robotic mirror therapy exoskeleton. This is to ensure that all people in need of such an exoskeleton would easily be able to obtain and afford one without struggle.

Cerebral palsy is the most common motor disability to occur in childhood often characterized with poor muscle movement and balance.

According to US studies, about two to three children out of every 1,000 are born with cerebral palsy averaging to about 10,000 babies born each year that will develop CP.

One type of unilateral CP is known as hemiplegia which is caused due to damage of the spinal cord or brain. Hemiplegic CP is usually characterized by paralysis on only one side of an individual’s body.

While there is no cure for a disorder such as hemiplegic CP, there are several ways that individuals can work to strengthen and gain better control of their affected muscles.

One way involves mirror therapy, a commonly used form of physical therapy utilized by recovering stroke patients.

While on a fundamental level some may consider mirror therapy to be meant for different types of physical therapy, it does show promise in helping individuals with hemiplegic cerebral palsy as it helps to improve the motor functions pertaining to the impaired side of their body.

However, simply simulating the motor function movement of an individual’s impaired side can only do so much. What CP patients need is a way to successfully train their impaired muscles through actual action, using repetitive, passive rehabilitation exercise.

Our team’s project for the STIP program involves creating a robotic mirror therapy exoskeleton that will drive the impaired arm of children with hemiplegic CP in response to the movement of their unimpaired arm.

The arm that our team has created allows for the planar flexion and extension of the participant’s elbow. The left arm is the “sensorized” arm that is being driven by the participant’s good arm, while the right arm is the motorized arm that mirrors the “sensorized” arm and moves the impaired arm with the help of a motor.

Both devices have potentiometers to sense the arm position, and a single microcontroller that bridges them together. The design of the device primarily consists of an armrest to hold the participant’s arm that would rotate at the elbow to extend the arm away from the user’s body. A hub under the elbow includes the necessary components to accomplish the rotation – bearings to provide smooth rotation, a potentiometer to sense the current arm position, as well as a motor in the slave device to acuate the motion.

An atmega328p based Arduino microcontroller was used to control the system of the exoskeleton. The motors were driven by a L298N H-bridge motor driver that has a limit of 48V with a peak current of 4A, and can be controlled by a 5V input signal. The motor driver also allows for current sensing, which can be utilized for overcurrent protection.

To assemble the arms together, our team used V-slot linear rails to allow for a flexible system where components such as forearm rests, handles, or other supports can be mounted to the arm assembly without having to redesign the entire assembly.

The control code for the microcontroller was written in C++. Running in a loop, the code would check the position of the potentiometers in both devices, check the current drawn to the motor, then calculate the desired input to the motor. If the desired position was outside of reasonable bounds, or the current to the motor was too high for too long, the motor would simply be stopped to prevent damage to the system.

Overall, the final design of the exoskeleton did function as was expected, successfully being able to move both arms together, mirroring the movements of each other. Our group had hoped to be able to add a wrist supination and pronation movement to the exoskeleton, but due to time constraints we were only able achieve the planar flexion and extension during the course of the summer. However, we do have CAD models demonstrating our group’s design for the wrist movement, showing that it is a viable option for future research, easily being able to be attached to our group’s current prototype of the exoskeleton thanks to the flexibility of the V-slot linear rails used for the arms design.

In the end, while our exoskeleton does successfully work and did have minimal costs, there is much that can be improved for future research. Both through design and types of materials, we can continue to improve and create more efficient exoskeletons for children with hemiplegic CP.

Summer Team Impact Project Undergraduate Research Scholars Program (URSP) - OSCAR

EKG based Heart Rate Monitoring and Reliability with Arduino

Author(s): Omar Alsuhaibani, Cynthia Love

Mentor(s): Nathalia Peixoto, ECE

We built a heart rate monitoring system with an Arduino microcontroller and ADS1292r analog-digital converter, that would allow real time monitoring of user heartbeat in a computer program. This computer program would also present live feedback to the user about their heart rate. One of the main aims of this project would be to explore if live biofeedback would help the user of the program control their breathing or heart rate, as well as test the reliability of the monitoring system, especially in comparison to the Empatica E4 wristband. Results suggested that the computer program monitoring heart rate in real time is similarly reliable to proprietary commercial monitors such as the Empatica E4 wristband.

Hello, we are one of the groups taking part in the STIP program for the summer of 2022. Our group consists of three members that worked on this research project over the summer. They are me, Omar Alsuhaibani, the group’s project manager German Kuznetsov, and Cynthia Love. The topic of our group’s project focused on attempting to use an Arduino microcontroller and an ADS1292r to create a system where a computer program can monitor the heartbeat of a user in real-time, and evaluate its effectiveness when the user is playing a video game.
Computer programs typically use devices like keyboards and mice to get input from the user. In our project we explore another form of input – biofeedback, primarily heart-rate, and test its reliability in different environments – such as when the user is sitting still, or playing a game on the computer.
While devices to get user heart rate in real time already exist, such as the E4, they do not provide direct access to the data. Because of this we chose to use an arduino with a ADS1292r sensor shield to monitor the user’s ecg signal and calculate their heart rate from the signal. The user will have two electrodes placed on their wrist, which will be connected to the ADS1292r sensor and arduino. The arduino will read the raw data from the sensor and send it to a PC for processing. The raw data signal goes through multiple filters in the computer program, which help remove noise from the signal and make it easier for the code to detect the heartbeats.
The result we reached is that the process is not 100% reliable. Even small movements by the user can disrupt the signal enough to where the computer program can no longer correctly identify heartbeats. Incorrect electrode placement or inadequate adherence can contribute to weakened connectivity.
However, since heartbeats are periodic and happen from around 40 to 200BPM, we can analyze the acquired heartbeats and make assumptions and conclude if they were detected correctly. By comparing the number of heartbeats detected correctly to those incorrectly, we can estimate the reliability of the acquisition method.
Using this setup, games and other computer programs will be able to incorporate biofeedback to measure the heartbeat and stress of the user. This can potentially be
used as a biofeedback visual that can aid the user in analyzing their heart rate and detect stress throughout a video game, VR simulation, or any other related activities.

College of Humanities and Social Science Honors College Summer Team Impact Project

Correctional Compassion: How does Empathy’s Association with age Influence the job Performance of Correctional Officers Working in Restricted Housing Units?

Author(s): Ella Tomita, Alexa Buckner

Mentor(s): Taylor Hartwell, Criminology

Currently, no research considers empathy among those working in restricted housing units (RHUs), which is also known as solitary confinement. Our research focuses on how empathy and other characteristics influence correctional officers in RHUs. The data used was collected from six male prisons and one female prison between 2017 to 2019. Ethnographic observations were the main method of data collection. 45–60-minute semi-structured interviews were conducted for 89 correctional officers. We coded 89 correctional officer interviews using ATLAS and used 73 of those interviews in the analysis. While coding we took note of concepts like age, positive or negative characteristics mentioned, interpersonal relationships, and the correctional officers’ definitions and perceptions of empathy. After coding, each interview was analyzed. The correctional officers were then grouped into the following age groups: young adults (aged 20 to 39 years old), middle adults (aged 40 to 59 years old), and older adults (aged 60 to 80 years old). Within their respective age groups individuals were placed in subgroups pertaining to their displayed empathy and positive or negative characteristics. The three subgroups were, “displayed empathy and positive characteristics”, “not enough information”, and “lacking a display of empathy and/or displayed negative characteristics”. We found that out of the three age groups, the correctional officers in the “Older Adults” group consistently displayed the most negative characteristics and lacked the most empathy compared to the other age groups. In the context of our research, the theoretical implication is that a person’s age may influence their empathy and job performance. Additionally, the practical implication is that older correctional officers displayed the least amount of empathy, which leads us to wonder if there should be an age cap for working in RHUs.

Correctional Compassion Script

Title Slide: (Alexa and Ella)
Alexa Introduction:
“Hello, my name is Alexa Buckner. I am going to be a senior this year at George Mason, and I am studying Criminology, Law and Society with a minor in forensic psychology.”

Ella Introduction:
“Hello, my name is Ella Tomita. I am going to be a sophomore this year at George Mason, and I am studying Criminology, Law and Society with a double minor in forensic psychology and social justice and human rights.”

Introduce our Project (Alexa):
“We are a part of a project that is aims to acquire a better understanding of the experiences of inmates, who we refer to as residents, and correctional officers living and working in solitary confinement units which is also known as RHUs or restricted housing units. For our research this summer Ella and I focused on how empathy and other characteristics influence correctional officers and their work.”

Previous Literature: (Alexa)
Transition/Introduction:
“While reading literature for our topic, we examined empathy broadly because there is a lack of literature examining empathy within the criminal justice system.”
First article:
“We first found that there is a negative association between age and empathy. This negative association is most often known to be caused by life experiences like physical disabilities, socioeconomic status, widowhood, and personal characteristics.”
Second article:
“Overall, within the research we found, Older adults have displayed an age-related decline in cognitive empathy and social functioning.”
Third article:
“We looked to other professions’ research on empathy and found that health care professionals who are more empathetic have been found to provide better treatment. This increased quality of treatment has been shown to increase patient compliance. Additionally, it was found that health care professionals who had interpersonal relationships have more empathy.”

Gaps in the Literature: (Ella)
“Currently, no research considers empathy among those working in restricted housing units. Consequently, there is no research that specifically addresses how the combination of empathy and age influence the ability of restricted housing unit correctional offices to do their job well.”

Research Questions: (Alexa)
Intro/Transition: “This gap in the research provided us with these research questions.”
First Question: “Our first question was if there are specific characteristics that lead correctional officers to be more empathetic towards residents.”
Second Question: “We also wondered if emotional empathy, (which is empathy that stems from sharing an experience, being willing to help other people, or experiencing distress in response to another person’s pain), or cognitive empathy, (which is empathy that stems from imagining what it would be like to be in another person’s shoes and understanding that person’s feelings), would be more important for correctional officers to display.
Third Question: “Our final question was what types of interpersonal relationships affect a correctional officer’s overall empathy.”

Data Collection: (Ella)
“The data used by our team was collected from 7 prisons – 6 male prisons and 1 female prison– between 2017 to 2019. Such data collection consisted of semi-structured interviews, which were about 45 to 60 minutes long, and ethnographic observations. During the data collection, interviews with 89 correctional officers were conducted.”

Data and Methods Slide One: (Alexa)
“We coded 89 correctional officer interviews using ATLAS. 73 of those interviews were used in the current analysis. This is a list of the majority of the topics or ideas within the interviews that we took note of while coding. Some of the most important here to note for the purposes of this analysis would be age, the positive or negative characteristics mentioned, interpersonal relationships, the correctional officers’ definitions and perceptions of empathy, the different types of empathy the correctional officers displayed, the correctional officers’ perceptions of inmates and their co-workers, and their willingness to report issues.”

Data and Methods Slide Two: (Ella)
“After coding, we analyzed the codes in each interview and then grouped the correctional officers by the following age groups: young adults (which consisted of individuals age 20 to 39 years old), middle adults (which consisted of individuals age 40 to 59 years old), and older adults (which consisted of individuals age 60 to 80 years old). Individuals were then grouped into three different subgroups, within their respective age groups, pertaining to displayed empathy and positive or negative characteristics. The subgroups consisted of displayed empathy and positive characteristics, not enough information, and lacking a display of empathy and/or displayed negative characteristics”

Examples of Categorizing Empathy: (Alexa)
“These are some examples of quotes from the interviews that we felt displayed empathy or a lack of empathy. One quote that we felt showed a correctional officers’ empathy reads, ‘At the end of the day, they’re still people (he is referring to the residents here).’ He goes on to say, ‘They deserve forgiveness… I’m not here to punish them anymore than their sentence.’ This next quote would be an example of a quote from an correctional officer that we found was lacking empathy. While discussing the mental health of residents in the RHU this correctional officer said, ‘ask me what bipolar is and I’ll tell you it’s a gay polar bear. I don’t care.’

Examples of Characteristics: (Ella)
“To preface, what we classified as positive and negative characteristics among correctional officers stemmed from what the correctional officers themselves said about themselves in response to questions throughout the interviews. Nevertheless, here are some examples of both positive and negative characteristics of correctional officers.”

“The first quote which states “Establishing a rapport with inmates is so important because every situation is unique. You need to know what every inmate needs in order for these units to keep running smoothly” demonstrates being prosocial with inmates, which is a positive characteristic. Contrarily speaking, an example of a negative characteristic among correctional officers is passiveness, which is demonstrated in an exchange where the interviewer asked a correctional officer, “Would you go to management if there were a problem with a coworker?” to which the correctional officer replied “No, I never would”

Findings: (Ella first, Alexa second)

Main Findings (Ella):
“Ultimately, we found that out of the three age groups, the correctional officers in the “Older Adults” group consistently displayed the most negative characteristics and lacked the most empathy compared to the other age groups.”

Categories (Alexa):
“Here within our three age groups, we have a breakdown of how many correctional officers were placed in each category after analysis. For the young adult age group which consisted of correctional officers between the ages of 20 and 39, 40% displayed positive characteristics and empathy, 35% did not have enough information, and 25% displayed negative characteristics and a lack of empathy. Within the middle adult age group which consisted of correctional officers between the ages of 40 and 60, 73% displayed positive characteristics and empathy, 10% did not have enough information, and 17% displayed negative characteristics and a lack of empathy. For our older adult age group that consisted of any correctional officer over the age of 60, 100% displayed negative characteristics and a lack of empathy. These findings aligned with the conclusions previous research about empathy have made.

Discussion: (Ella first, Alexa second)

Limitations (Ella):
“One of the limitations of our research was the availability of interviews with Older Adults. In other words, we had well over 20 interviews for the other two age groups, but only 4 with the Older Adults”

Theoretical Implications (Ella):
“The theoretical implication in the context of our research is that a person’s age may influence their empathy and job performance.”

Practical Implications (Ella):
“Additionally, the practical implication in the context of our research is that older correctional officers displayed the least amount of empathy, which all together leads us to wonder if there should be an age cap for working in restricted housing units.”

First Next Step (Alexa):
“The first step we would like to take next regarding our research is to investigate whether correctional officers nearing the older adult age bracket begin showing signs of a decline in empathy or social functioning. We would also like to code resident interviews to investigate their perceptions of positive correctional officer characteristics. Additionally, we could do a mixed methods longitudinal study to see if individual correctional officers’ empathy and social functioning changes overtime and with aging.”

Conclusion (Alexa):
“Thank you so much for your time, and for listening to us today.”

Summer Team Impact Project

Curating the Chinese Anti-Rightist Campaign of the 1950s

Author(s): Joel Adeniji, Bailey George, Kristie Phan, Weilon Price

Mentor(s): Shanshan Cui, Art and Design Building

Curating the Chinese Anti-“Rightists” Campaign in the 1950s

The Chinese “Rightist” Archive is a collection of testimonials, biography, and images of intellectuals who were accused for being against the government during the Chinese Anti-“Rightist” movement between 1957 and 1959. While there is great potential of large-scale analytics and historical research in the archive due to over 40,000 documents, the unstructured nature of the collection and their data format of plain text make it difficult for users to access necessary information. Our 2022 Summer Team Impact team aimed to make this collection accessible and searchable by leveraging computational archival science (CAS) methods and design methodology from the field of Human-Computer Interaction (HCI). Specifically, we (1) developed automatic data pipelines to structure, translate, and standardize the raw data using Python, Amazon Web Services, and OpenRefine, (2) designed a real-time database schema to ensure data management efficiency and searchability, (3) designed the user interfaces and user experiences using the scenario-based design methodology, and (4) implemented the web-based archival platform using Firebase, Angular, and other cloud-based APIs. The final application not only resolves the information accessibility issues by providing searchability and navigability of the data, but also makes it possible to add additional data by providing public contribution pages. This research can contribute to CAS with the data processing pipeline for language translation and structuring and to HCI with the adjusted design methodologies in the context of individual-level historical data.

00:01
Hello, we are the summer 2022 impact team for the project, Curating the Chinese Anti-Rightists Campaign in the 1950s. For our project, we created a website that focuses on the Anti-Rightist Campaign through our research and data we currently have in our database.

00:16
To give some background information about the Anti-Rightist movement, it began in 1957 which lasted for two years from 1959. The Chinese Communist party reeducated, imprisoned, and executed those who were counter-revolutionaries of the government. For our research, we have 40,000 files in our database to include in our project website which contains rightists who were impacted by this movement. While 40,000 may seem a lot, this is only a small portion out of the millions of lives that were affected by this campaign.

00:45
The research problem of our project contains 3 components. One, the information accessibility contains the 40,000 files we have in our database. So our team needs to figure out how to implement that factor into our website to help users search for specific files. Two, the data management contains unstructured data which makes it hard for the practitioners to manage and organize the data. Finally, the design problem is that even if the data are well organized, it is hard for users to use the data intuitively, so we need to use UX/UI design to help improve usability in the system.

01:19
To briefly explain our approach in the CAS also known as the Computational Archival Science, the problem lies within the information accessibility and data management. We have to create easy navigation and searchability through digitizing the archival collection while structuring the data using Python scripts and managing the data in the firebase database. Finally, the design problem lies within the human computer interaction, or HCI, which uses a scenario based methodology to generate requirements and usability in the system.

01:50
During the beginning of our project, we brainstorm some of the timeline we need to take for us to view the application and some of the architecture and the schema aspect of it.

02:03
We recurate through a lot of the database, so we develop Python scripts to translate the raw files into a more schema format.

02:16
We use Angular, Github, and Firebase for setting up the front end aspect of it to use Angular and we use Github to maintain and manage the code base. And we also use Github to set up the workflow for automation view and automation testing deployments. And for the firebase, we use this to all the applications both the authentication and the domain aspect of it.

02:46
So, here are some of the wireframes the design team that created for the interface and database design once the user scenarios were tested.

02:55
Here are the overall final designs of the website and database that were just completed.

3:04
Here’s a close up of the archive collection views and the rightist file page. The two on the left showcase our most important feature on the website, which is the ability to search the entire database for a rightist.

03:17
And then, here are some close ups of the account pages for a logged-in public user which displays their contributions shown on the left and on the right, if they were to upload a new rightist file for the database.

03:33
For the Computational Archival Science, we basically developed a new data processing pipeline for unstructured Chinese data.

03:43
And we are going to move on with the DEMO of the website.

03:48
When you first come to the website, this will be the homepage. We have a search bar right here, which will allow the user to search for any of the documents.

03:58
The basic overview of the homepage and the users are also going to switch between Chinese and English, like so.

04:08
Now let’s move on to the login functionality. Users can register with Google or Facebook or login with Google. I will login to Google in this case.

04:22
And users are also able to upload additional files that contribute to our existing database. And we are able to search this new contribution in our search functionality. By the way, here we have the archive of all the other individuals.

04:43
And this is our project and those are the main functionality. Of course, we have other pages the user can visit and I’ll just leave that for you guys to explore, so thank you.

College of Science Summer Team Impact Project Undergraduate Research Scholars Program (URSP) - OSCAR

Music and Emotions

Author(s): Omar Ahmed Alsuhaibani, Pamela Benitez, Cynthia Love

Mentor(s): Nathalia Peixoto, Bioengineering

https://youtu.be/oK0Bic28M-k

This study explored the efficacy of computer-generated music in comparison to human-composed (commercial) music among students and examined how music tempo and style related to changes in physiological signals such as heart rate and EEG signals from the brain. Participants were 6 students ranging from ages 15 to 22 years old, both male and female. During the time of testing, participants were given ten six-minute tests: one as a baseline reading of physiological signals, three tests involving commercial or human-generated music of varying tempos, and 6 tests involving the computer-generated music at different speeds and styles. Results suggested that computer-composed music is (a) positively correlated with a change in physiological signals, (b) more effective than human-composed music, and lastly that music tempo had a (c) positive correlation with gamma and beta brain waves. Results are discussed in terms of implications for students’ changes in heart rate and EEG/brain wave activity.

Hello, we are one of the groups taking part in the STIP program for the summer of 2022.
Our group consists of three members that worked on this research project over the summer. My name is Cynthia Love, serving as the project manager. I am also working alongside Omar Alsuhaibani and Pamela Benitez. The purpose of our group’s project was to explore how different styles, sources, and speeds of music affect physiological signals such as EEG signals from the brain or heart rate.
Some real life applications are to gain the ability to measure emotions, something that has not yet been easily quantifiable. Another application is to optimize an up and coming field of therapy known as music therapy, especially for those with auditory triggers. While music therapy has grown in popularity for its ability to effectively reduce feelings of anxiety and stress, it was unknown if computer-generated music was as effective as music composed by humans. This was another research question we had in mind during our experiments.
Our group hypothesized that there would be a positive correlation between music tempo and beta and gamma brain waves. We hypothesized that fast music would enhance beta and gamma brain waves, while slow music would suppress them. Our last hypothesis is that computer-generated music would be less effective at influencing physiological signals.
Materials – The equipment used for our experiments is listed on the screen. The muse headband was used to take EEG brainwaves, while the empatica wristband was used to record heart rate. The mind monitor app allowed us to process the raw EEG data gathered from the experiments. The E4 connect software was used to process the heart rate data. The Interamusi Software is where we derived most of our music from. It is a website that composes custom music and allows it to be played on a loop.
Testing – this is the type of software that we used for each data processing. The one on the left is the empatica real-time app, which was used to process heart rate data. The app on the right is the mind monitor app, which was used to process EEG data.
Methods – For this experiment, we first started with a baseline test. In this portion, participants were instructed to sit in an empty room, with their eyes closed. Lights were turned off in the room to prevent any electrical interference with the equipment used. The Muse 2 headband was placed onto the participant, as well as the E4 on the participant’s wrist. Both devices were turned on and captured the participant’s readings for 6 minutes. The next portion involved the music tests. Participants listened to each of the 9 computer-generated songs for 6 minutes each. These songs were split into two categories, computer-generated classic music, and computer-generated modern music. There were then three subcategories, known as calm, stimulus, and focused. The order in which the music was played was randomized for every single participant.
For each stimulus, the Muse 2 headband and E4 were activated at the start of each 6-minute trial. Each six-minute session was recorded through the Mind Monitor app and labeled correspondingly to the stimuli played.
For the baseline test, the four different channels both dropped leading to a decrease in Power Spectral Density with no brainwaves appearing for delta from 0.5 to 3 Hz and theta frequencies from 3 Hz to 8 Hz. This confirms the accuracy of the results interpreted as delta waves are generated from the deepest form of meditation and dreamless sleep, while theta waves are generated from learning and intuition and can be thought of as generated from beyond a conscious level. We can see an increase in activity from 8-12 Hz which indicates that there is minimal alpha waves activity occurring. Alpha waves are the present resting state of the brain where wave presence indicates relaxation and mental coordination. Frequencies of 12 Hz to 32 Hz are considered beta waves which are indicators of normal brain activity such as having our eyes open and processing information on our surroundings. In regards to our baseline testing, there are some beta waves present in the different channels which can indicate that the test subject was processing the experiment and the environment during testing. The highest activity present in the baseline testing were gamma waves. Gamma waves occur at 32 Hz and above and are considered the fastest waves out of all the aforementioned waves. Gamma waves are associated, and mainly occur, with high levels of thought and focus when conscious. In this baseline test, there were high spikes of gamma waves which indicate most likely that the subject was focused on staying still and clear headed.
With an overlook of the slow commercial music channel spectra, we can see that there is much more activity in PSD and rise in waves for different frequencies. During 3 – 8 Hz, there is no noteworthy activity for theta waves from all channels. This indicates that there was little to no learning or intuitive activity beyond a conscious level during the listening to slow commercial music test. In the 8 – 12 Hz range, there is minimal activity for alpha waves for all channels. This indicates that there was minimal mental coordination and relaxation during testing. For the 12 – 32 Hz frequency range, there is significantly higher PSD and waves from one channel than the others for beta waves. This could indicate that one area of the brain picked up more beta waves than the other channels, which is an indication that there was higher level of concentration on processing the environment they were in. In the 32 Hz and above range, there is significantly higher activity for all channels in regards to gamma waves. This indicates that the slow commercial music resulted in a high level of conscious thought and focus.
At a first look at the classic calm computer-generated music channel spectra, we can see that there is a higher activity in all the channels throughout the frequency range and PSD. During 3 – 8 Hz, there is little activity for theta waves from half of the channels but the PSD does start at a lower value than the commercial music test for all channels. This could indicate that there little learning or intuitive activity beyond a conscious level during the listening to classic calm computer-generated test. In the 8 – 12 Hz range, there is the activity for alpha waves for all channels with some channels slightly dropping in PSD while other channels have alpha waves forming. This indicates that there was minimal mental coordination and relaxation during testing. For the 12 – 32 Hz frequency range, there is a rise in PSD and waves from two channels for the beta waves. The two areas of the brain picked up more beta waves than the other channels.
Specifically, for one of the channels at low Beta (at 12-15 Hz) there is a slight formation of low Beta waves that remained consistent until it reached Beta (15-22 Hz) and higher activity was detected through the appearance of multiple Beta waves. The same can’t be said for the other two channels, where when it reached Beta, there was no increase of beta activity due to no waves appearing and the PSD slowly decaying. This can be interpreted that there was thinking occurring during testing with an awareness of self and focus on surroundings during Beta frequency picked up in two areas of the brain [NHAHealth]. In the 32 Hz and above frequency range, there is significantly higher PSD and wave activity for half the channels that proceeded to all channels when it reached above 50 Hz, in regards to gamma waves. This indicates that the classic calm computer-generated test resulted in a high level of conscious thinking and a high level of processing that is more consistent and organized, due to the firing rate of gamma waves, in all areas of the brain during testing.
In conclusion, this project achieved to explore the efficacy of computer-generated music in comparison to human-composed music among students and examined how different music styles and tempos can relate to changes in physiological signals such as EEG brainwave signals. The computer-generated music had a random pattern of behavior on the brain with the four areas of the brain. With the computer-generated music tests, after analyzing the data acquired, the four different channels were picking up the same signal and matching each other in regards to activity. This happened with human-composed music, as well, but not as often or frequently as during computer- generated music. This gave more solidity to the findings and the interpretation of the graphs. It proved that there was higher and stronger activity during particular frequencies picked up from all areas of the brain that can determine how strongly the subject was feeling. The purpose of this project was to examine if human-composed music and computer-generated music can have the same implications on human emotions and physiological signals. The results proved that there is a difference based on signal activity, PSD, and frequency of waves but more testing and analysis are required to reach a fully-formed opinion.

College of Engineering and Computing College of Public Health Summer Team Impact Project Undergraduate Research Scholars Program (URSP) - OSCAR

Personalized Intervention Using Personalized VR Recovery Rooms and Avatars

Author(s): Omar Alsuhaibani, Taylor Dinh, Elizabeth Kwari, Sara Razavi

Mentor(s): Nathalia Peixoto, Bioengineering

This project focuses on creating a personalized virtual reality (VR) safe space and avatar for individuals that are recovering from substance use disorder. HTC Vive (VR headset), Unity (Game engine), VRoid (Avatar assembly and characterization), and Blender (3D-design software) are used to create the VR safe space and avatar. The HTC Vive is calibrated and linked to Unity, and eye tracking scripts are implemented in Unity. To test the effectiveness of the VR safe space and avatar, an E4 is connected to the computer and to Unity to record the heartbeat in real-time. The heartbeat is then translated into the avatar’s animation using VRoid to simulate the emotion during particular heart rates as an indication of stress, anger, or fear.

(Slide 1)
Hello we are one of the groups taking part in the STIP program for the summer of 2022. The topic of our group’s project focused on creating a personalized VR recovery room and avatar for individuals that are recovering from substance use disorder (SUD).

(Slide 2)
And this brings us to our group’s overall goal statement, a reiteration of what was previously stated. Our group wanted to create a personalized VR safe space and avatar for individuals recovering from substance-use disorder.

(Slide 3)
The reason we decided to do more research pertaining to the topic of VR and personalized intervention has to do with the various forms of literature that are in support of VR technology as a useful method for recovery therapy.

For example, there was one study where VR was used on patients & healthcare providers in COVID-19 rehabilitation units. They found that VR helped them to cope with the isolation and loneliness they felt while quarantining during the beginning of the COVID-19 outbreak.

Another example can be seen when patients in mental health care faculties used VR technology. It helped individuals that experienced high levels of stress reduce their stress levels and maintain a calmer attitude in their daily life.

Finally, in a study that was conducted on drug abusers in a drug rehabilitation center in China, researchers found that after VR treatment, the participant’s craving for drugs was severely reduced.

For this reason, our group felt that looking more into and trying to personalize VR safe spaces and avatars would be a viable and useful method to assist individuals recovering from substance use disorder.

(Slide 4)
The design of this project was created in a temporary manner by using HTC Vive, Unity, and Blender for the development of the personalized recovery room.

(Slide 5)
Hi, my name is Taylor Dinh, I was the programmer on this project. The video you can see here is a recording of the viewport and project windows within Unity. I programmed in Unity, a game engine which runs off of C#, in order to create a VR room that can be viewed and interacted with through the HTC Vive and controllers. Our original intention was to make a therapeutic room where people with substance use disorder could go into the room and see an avatar that would mimic their heart rate using the animation controller and an Empatica. Unfortunately, we could not get this to work, however, we did make a VR room that the participant can interact with.

(Slide 6)
For the Empatica set-up, E4 streaming desktop app was used in order to keep track of the real time data streaming from the device attached to the participant using a bluetooth dongle connected to the computer
The BLE github code to connect Unity to the E4 was used in order to attempt to connect the two so that the player could see their metrics in real time

(Slide 7)
In terms of the personalized avatar, our team had initially started constructing the avatar using Blender. However, it proved to be much more challenging than what had initially been expected. While it is a viable solution, it would take time and a lot of effort to construct a model personalized to an individual’s preference.
Instead, our group looked into personalizing avatars using VRoid. This method proved to be much easier in terms of personalizing the avatar compared to building it from scratch in Blender. Creating the avatar proved to be a lot easier in VRoid as it includes several type of customizable features and can be easily imported into Blender and then Unity which is very helpful. If you are interested in learned more about creating your own personalized avatar and importing it into Blender, you can check out the full version of this video on our teams website.

(Slide 8)
In conclusion, the VR personalized intervention room and avatar that this paper has attempted to develop went successfully in building a minimal demonstration of the capabilities VR technology can have. The VR personalized intervention room model developed here integrates highly important aspects, such as head tracking and eye-tracking, to ensure that the user focuses on the intractable objects around them. In addition, the model developed had the ability to present generic soothing images in an act to calm down the users.
The purpose of this paper was to demonstrate an inexpensive method to develop a personalized recovery room and avatar with the use of VR based on heart rate.
The results were relatively similar where the personalized recovery room was able to become developed, albeit not personalized to particular users and their needs. However, the results prove that the ability to develop personalized recovery rooms that implement important personal variables (such as the users’ dogs, family homes, or family pictures) is a lot more difficult and requires a lot more time by trained professionals.

College of Humanities and Social Science College of Science Summer Team Impact Project Undergraduate Research Scholars Program (URSP) - OSCAR

Building Capacity for Indigenous Mapping & Data Sovereignty: In collaboration with the Chickahominy Indian Tribe and the Upper Mattaponi Indian Tribe

Author(s): James Condo, Adam Edwards, Domi Hannon, Sara Jefferson, Brian Jimenez, Paloma Jimenez, Maiya Justice, Guadelupe Meza-Negrete, Jasmine Okidi, Patricia Troup

Mentor(s): Jeremy Campbell, Institute for Sustainable Earth; Tom Wood, School of Integrative Studies

The Indigenous Environmental Mapping and Resilience Planning Project has worked over the summer with two of the federally recognized Native American tribes of Virginia, the Chickahominy Indian Tribe and the Upper Mattaponi Indian Tribe. As mentioned in our first abstract, Western academia has participated in an exploitative, extractive relationship with Indigenous tribes. Our project challenges the structures of settler society by aligning our research objectives with goals and values provided by the Chickahominy and Upper Mattaponi. These objectives include building tribal capacity in mapping and data sovereignty to help inform tribal environmental decision making. Honoring Indigenous knowledge is key to our project’s overall goal of Indigenizing academia, including honoring Indigenous data sovereignty. Thus not all aspects or results of our project are shown in this video, in accordance with the wishes of the Chickahominy and Upper Mattaponi. Project deliverables include a GIS training workshop for Chickahominy tribal environmental managers, a preliminary GIS database of Chickahominy and Upper Mattaponi tribal land and bioculturally significant sites, a visualization of landscape change within Upper Mattaponi tribal lands, a literature review on Indigenous data sovereignty, and a preliminary data analysis of demographic and environmental questions from the Upper Mattaponi Tribal Wellness survey

[Two undergraduate researchers stand against a white wall, and assorted photos and video from Chickahominy and Upper Mattaponi fieldwork]
Maiya: Our project, the Indigenous Environmental Mapping and Resilience Planning Project, has been working over the summer with two of the federally recognized Native American tribes of Virginia – the Chickahominy and Upper Mattaponi Indian Nation. Our work with these two tribes has been to establish environmental data baselines, using specific guidelines given to us by the tribe’s environmental offices.
Paloma: As we’ve mentioned in our previous video, maintaining tribal sovereignty over their respective data and knowledge is key to our work. This video will focus primarily on Indigenous Data Sovereignty, the GIS Map training and work we’ve been conducting over the summer, and our time with the Upper Mattaponi Indian tribe.

[Footage and images from the Upper Mattaponi PowWow interspersed with images and footage from Upper Mattaponi lands as well as GIS footage]
James: Indigenous data are defined as any piece of information that impacts Indigenous people as a group or as individuals, Indigenous lives, Indigenous cultures, and Indigenous languages (NCEAS, 2021). For example, Indigenous data may include environmental data such as geospatial or watershed information, data regarding Indigenous nations as a polity, and data about individuals such as demographic or epidemiological data (ibid.). Indigenous data are especially relevant regarding Indigenous sciences that leave further room for subjectivity and qualitative data, as well as the dissemination of Indigenous knowledge through oral histories. Sovereignty is broadly understood as rights of self-determination and self-governance without outside interference. In the context of sovereign Indigenous nations, Indigenous Data Sovereignty refers to Indigenous ownership and control of data, rather than data being owned by an outside university or government. Indigenous Data Sovereignty is critically important to enabling Indigenous environmental stewardship, collective knowledge ownership, and non-extractive data collection (NCEAS, 2021). There are many methods through which Indigenous leaders and collaborating researchers have worked to ensure long-term Indigenous data sovereignty and governance of information about Indigenous lives. Our project this summer has worked to establish and maintain IDS by deferring to leadership for the Upper Mattaponi and Chickahominy tribes in regards to data that is confidential. This has involved flexibility with internal use and delivery of data in which control lies in the hands of tribal leadership. While our team has worked to collect data, tribal leaders are able to govern the data that is relevant to the lifeways of tribal citizens.

[Footage of the Chickahominy GIS training session interspersed images and footage from Upper Mattaponi fieldwork and GIS footage]

Adam: Using GIS can be difficult to use, because GIS software has many capabilities and numerous ways to navigate the many menu options. GIS is Important to learn, because so much data today has a spatial component, and spatial problem-solving can be used in nearly every field. GIS is an important tool for analyzing important issues facing the earth today. For the GIS training with the Chickahominy Tribal members, ArcGIS Pro was used. ArcGIS Pro is a GIS software that allows you to access, explore, visualize, create maps, and analyze data geospatially. The most enjoyable part of the GIS training is showing how maps can be used to explain spatial relationships.

[Leigh stands in front of a wooded ponded area interspersed with GIS time series analysis]
Leigh: My name is Leigh Mitchell. I’m the environmental and cultural protection director with the Upper Mattaponi Indian Tribe, citizen of the Cherokee Nation. It is up to the tribe, uh, themselves to to lead, um, lead a project in a way where again their control and their use of their identity is up up to them, um, to describe, disseminate, research, um, look into, um, so while an individual tribal citizen can, um, of course you know interact and create data and kind of the the tribal sovereignty aspect is a whole new layer, um, that’s really important. Like I said kinda every step of the way we are considering,um, you know are we putting information at risk are we ensuring that, um, the information we want to be out there is out there are we sharing too much, um, how do we keep, um, internal knowledge internal, um, so when looking at environmental restoration I can I can say you know this is a sacred site or um or I can say there’s a sacred relative or a species there that is important, um, but any more you know that’s really up to the tribe to say what they want to say, um.

[Footage from the Upper Mattaponi PowWow]

Sara: At the beginning of the summer, some of our members had the chance to attend Upper Mattaponi’s yearly powwow, which had not been held since 2019. Here are some of the team members pictured with Upper Mattaponi Chief Frank Adams, enjoying the Powwow.

[Footage of Upper Mattaponi lands interspersed with footage and photos of some of the team meeting with Upper Mattaponi members]

Domi: While we weren’t able to conduct a wildlife survey in a similar vein to the one done on Chickahominy lands, we were able to meet with Leigh Mitchell, Environmental and Cultural Protection Director of the Upper Mattaponi Indian Tribe, and Reggie Tupponce, Tribal Administrator of the Upper Mattaponi. We were able to view some of the Upper Mattaponi lands, as well as the environmental risks in the area, such as a kitty litter factor and extensive sand and gravel operations. Some of our members worked on doing a preliminary data analysis using data from a Tribal Wellness Survey conducted by the tribe in 2021; data from this survey contains valuable insights into Tribal Members’ observations regarding the increased intensity of storms and droughts in the region. Our team has also partnered with Upper Mattaponi on important mapping projects, including the documentation of recently reacquired tribal properties using ArcGIS and Google Maps. Some of the featured data that we’ve developed include images of landscape and watershed change over time, as well as changes in air quality from 1998-2020. We’ve worked with Tribal Members and administrators to integrate this information into an accessible and secure database that can inform tribal decision-making for years to come.

[Three undergraduate researchers stand against a white background, interspersed with footage from Chickahominy fieldwork]

Domi: The work we’ve been doing this summer is a part of efforts to decolonize our work here at George Mason. What we’re doing isn’t about controlling research objectives, data, or access to data, but capacity building with Native tribes of Virginia to assist them in managing their own lands and cultures.
Guadelupe: We hope you’ve gained an understanding on valuing Indigenous Sovereignty and knowledge in academia and building and maintaining a reciprocal, additive relationship with Native peoples across Turtle Island and the world.
Patti: The relationships we’ve been building with the tribes are just as important as the data we’ve been working on. We would love to expand our work to other tribes in the Chesapeake Bay area, and that this sets the foundation for future work between George Mason and the Native folk whose land we continue to occupy.

[Thanks and credits set against PowWow footage as well as team photos and footage of both Chickahominy and Upper Mattaponi lands as music from the Upper Mattaponi PowWow plays]

College of Humanities and Social Science College of Science Summer Team Impact Project Undergraduate Research Scholars Program (URSP) - OSCAR

Supporting Tribal Sovereignty & Environmental Governance: In collaboration with the Chickahominy Indian Tribe and the Upper Mattaponi Indian Tribe

Author(s): James Condo, Adam Edwards, Domi Hannon, Sara Jefferson, Brian Jimenez, Paloma Jimenez, Maiya Justice, Guadelupe Meza-Negrete, Jasmine Okidi, Patricia Troup

Mentor(s): Jeremy Campbell, Institute for Sustainable Earth; Tom Wood, School of Integrative Studies

The Indigenous Environmental Mapping and Resilience Planning Project has worked over the summer with two of the federally recognized Native American tribes of Virginia, the Chickahominy Indian Tribe and the Upper Mattaponi Indian Tribe. Historically, Western academia has participated in an exploitative, extractive relationship with Indigenous tribes. Conversely, our project challenges the structures of settler society by aligning our research objectives with goals and values provided by the Chickahominy and Upper Mattaponi. These objectives include establishing environmental data baselines for their ancestral lands, such as recognition of Indigenous food sovereignty and biodiversity mapping. The Indigenous Environmental Mapping and Resilience Planning Project is grounded in collaborative practices and support for Indigenous environmental stewardship while upholding Indigenous data sovereignty. This project is the first step in establishing long-term reciprocal relationships between George Mason University and the Chickahominy and Upper Mattaponi. Project deliverables for the food sovereignty portion include an Indigenous gardening booklet, a presentation on composting, and signage for a community garden. Project deliverables for the wildlife assessment portion include an ethnobiology booklet covering bioculturally significant species and a breeding bird survey and an invasive species survey conducted on recently repatriated Chickahominy ancestral lands.

Dana: Okay the tribe is headquartered in Charles City County, Virginia at an area known as Chickahominy Ridge. My name’s Dana Adkins, I’m a Chickahominy Tribal Citizen and the tribe’s environmental director since 2019.

Sara: Soo my name’s Sara Jefferson, I’m a member of the Chickahominy Indian Tribe, and I’m also a student here at Mason studying psychology, and I’m also a team member of this STIP project going on here down in Charles City.

Dana: Some of the deliverables we are hoping to get from this project are maybe introducing us to some methods that can help us to see what lands we have, understand what’s there, what’s not there, and how to best utilize those lands so that as we grow, we can have growth that’s environmentally friendly.

Sara: To me, sovereignty means kind of taking back what was already ours, if that makes sense, being that we were one of the first tribes to have direct contact with the colonizers back in 1607, and so much of our culture was lost and we weren’t able to get all of that back, but now that we are a sovereign nation, we are able to get these tools that we once lost back in an easier manner, if that makes sense.

[A group of four undergraduate researchers are standing against a white wall]

Domi: Our project, the Indigenous Environmental Mapping and Resilience Planning Project, has been working over the summer with two of the federally recognized Native American tribes of Virginia – the Chickahominy Tribe and Upper Mattaponi Indian Tribe. Our overall goal has been to establish environmental data baselines for the two tribes, using specific guidelines given to us by the tribe’s environmental offices.

Guadelupe: Since the beginning of our relationships with the tribes, we’ve focused on maintaining tribal sovereignty over data and knowledge, as historically, Western academia has participated in an exploitative, extractive one-sided relationship with Indigenous tribes and their knowledge across North America, also known as Turtle Island.

Patti: Having a reciprocal relationship built on trust and honoring Indigenous knowledge is key to our project – we want our work with the Chickahominy and Upper Mattaponi to be additive and beneficial to the tribes. By participating in this work, we are attempting to decolonize academia and environmental studies.

Adam: An important part of our research has been honoring Indigenous data sovereignty – that means not everything we’ve worked on over the summer we can show, as the tribes do not want all data available for access to non-tribal members.

[Footage of the Powhatan River]

Dana: The site we’re standing on now is a piece of property that we have named “Chickahominy on the Powhatan.” Uh, this would have been a traditional village site, uh, the rivers uh in the state served as great village sites for the tribes. Uh, the tribes in this state are actually known as tributary tribes because of our villages being along our namesake rivers. Uh, this particular river would have been called the Powhatan River, pre-contact. Uh, post-contact it was renamed the James River.

[A single undergraduate researcher stands with a glass door behind her, with different screenshots interlayed over her at different points in time]

Jasmine: Indigenous and settler scholars Eve Tuck and K. Wayne Yang explore how the metaphorization of the term “decolonization” has strayed the operational form of the word away from the objectives of Indigenous rights and sovereignty. This metaphorization may look like conflating decolonization with other social justice movements. This metaphorization may also manifest in “settler moves to innocence,” or methods of alleviating settler guilt from Indigenous erasure and forced assimilation. This is achieved by creating distance from the settler to the system of settler colonialism without actually absolving themselves of oppressive power or returning stolen land. Rather than falling into these inappropriate comparisons or settler moves to innocence, it is important that we recognize decolonization as a distinct project on the repatriation of Indigenous land and lifeways. For our project, that means challenging the structures of settler society by supporting Indigenous environmental stewardship and ensuring Indigenous sovereignty over data and, crucially, guiding our work with the stated goals and values of the respective tribes with which we are working.

[Three different voice overs over footage of Chickahominy lands as well as the Potomac Heights Community Garden]

Paloma: Over the summer, we have been working closely with Dana Adkins, the Tribal Environmental Director of the Chickahominy. He’s been working on establishing community gardens for the tribe to ensure healthy living for tribal citizens. With Dana, we’ve been working on establishing food sovereignty for the tribe through these community gardens as well as creating a curriculum on how to teach food sovereignty and the issues of food deserts and accessibility.

Patti: Food sovereignty is the right for peoples, especially native peoples, to have access to healthy, environmentally sound, traditional food. This includes both the ability to cultivate, harvest, and consume those foodways. Food sovereignty does not listen to market demands, but instead, focuses on the need of the people of the respective culture. Food sovereignty is being used across Turtle Island and worldwide to combat food deserts. Food deserts are when an area has a rate of poverty exceeding 20%, with rural folks being unable to access a large grocery store within ten miles of their home. The area in which the majority of the Chickahominy people live in is considered a food desert. The community garden and eventual future gardens established by Dana will help combat the lack of access to healthy, in-season food.

Maiya: One of the ways our team has worked on the community gardens is through research at George Mason University’s own community gardens, maintained by the Office of Sustainability. Guided by Doni Nolan, the Sustainability Program Manager for the Gardens and Greenhouse, we gained insight and ideas to present to Dana on working on the garden to best benefit the citizens of the Chickahominy nation.

Dana: Well we received a grant from the EPA this year to start a community garden. We as a tribe have gotten away from gardening, as a general rule. Like most of society we live fast paced lives and we rely more on fast food and prepared food, rather than stuff than we would have cultivated ourselves out of our gardens and cooked at home. The thought behind having the community garden was to kind of reintroduce some of those gardening techniques to our youth before it’s lost. We also have, like other communities, we have seen an increase in diseases such as diabetes and we feel this is probably attributed, again, to too much fast food, not enough of the foods that we grow ourselves, that are generally healthy, uh foods that you’d want to consume.

[An undergraduate researcher walks along a green path and his voice is overlaid with footage from Chickahominy lands]

For the Chickahominy, we also began collecting data on the wildlife on the lands that currently have regained ownership over. The two lands, Chickahominy on the Powhatan, or James River, and Mamanahunt on the Chickahominy River, were both recently (re)acquired through purchase by the Chickahominy. We participated in a bird survey, led by Dr. Wood, and collected data on the birds, other animals, insects, as well as the flora growing on the Chickahominy properties. We also conducted basic water monitoring from the Powhatan, or James River, and the Chickahominy River as well. One of our deliverables will be an ethnobiology booklet focusing on a select number of native flora and fauna.

[Footage from the Upper Mattaponi Powwow of Women’s Fancy Shawl Dancing]

Domi: Our work with the Chickahominy tribe of Charles City, Virginia in food sovereignty and wildlife data collection is only part of what we’ve been doing this summer – check out our second video to learn more about our mapping efforts with both the Chickahominy and the Upper Mattaponi Indian Tribe!

College of Humanities and Social Science College of Science Summer Team Impact Project

Role of the Urban Built Environment in Breast Cancer Mortality Health Disparities

Author(s): Kai Barner, Ha Dao, Abigail Kokkinakis, Alexandra Merida, Amanda Webber

Mentor(s): Taylor Anderson, Geography & Geoinformation Science; Travis Gallo, Environmental Science and Policy; Mariaelena Pierobon, Biohealth

Breast cancer is the second-leading cause of cancer-related death among women in the United States, but mortality rates vary across the population. Previous studies use individual and county level data to examine the association between breast cancer mortality and various socio-demographic and environmental variables. However, global regression models assume spatial stationarity meaning that associations between explanatory variables and breast cancer mortality are the same across geographic space and scales. Therefore, the objective of the study is to use county-level data that describes the social and built environment across the contiguous United States to explain breast cancer mortality rates, as reported in the Surveillance, Epidemiology, and End Results (SEER) database, using a multi-scale geographically weighted regression model (MGWR) that accounts for spatial heterogeneity. We compare our MGWR model with a baseline global linear regression model (OLS). The MGWR outperformed the global linear regression model with an adjusted R2 of 0.91, explaining about 43% more variance than the OLS. The R2 for each county is high in the western part of the US and decreases towards the east. In comparison to global linear regression, MGWR had the same overall trends in terms of relationships between explanatory variables and mortality. For example, as mammogram screenings, health food index, and primary healthcare physician ratio increased, breast cancer mortality decreased. However, MGWR reveals spatial heterogeneity associated with the magnitude and direction of each relationship across counties. Such an approach allows for greater consideration of where certain variables are most influential in breast cancer mortality allowing for location-specific interventions.

Top of Form

Abigail G Kokkinakis0:09

Hello everyone. The objective of this study is to examine the association between the urban built environment and sociodemographic variables and breast cancer mortality using an approach that will capture spatial heterogeneity and those associations. According to the American Cancer Society, breast cancer is most common cancer in American females. One in eight females will be diagnosed with breast cancer and their lifetime, and one in 39 females will die from breast cancer. A lot of studies look at individual level, social demographic data, and regression techniques. However, these regression techniques assume spatial stationarity which ignores local heterogeneities and associations between variables and breast cancer mortality.

Ours created a US model focusing on urban built environment variables without the limitations of most global regression techniques.

We collected Seer data from the most recent five-year interval, which was 2015 to 2019. Note the white space in the map shown. These are the counties that did not have data provided, which was roughly 1/3 of the data.

Ha Dao1:10

Due to privacy reasons, counties with the count of deaths less than 10 were omitted by SEER. Therefore, we decided to assume these missing counts as 9 and calculate the crude rate based on this. Moving forward with data collection and preprocessing. In the beginning of our research, 65 urban-built and social-demographic breast cancer determinants were obtained. Throughout the process of data cleaning and data wrangling, log transformation and scaling were applied on selected variables to reduce skewness and normalize the range of independent variables. Upon running Pearson correlation and Variance Inflation Factor, we removed variables that were highly correlated so we can avoid multicollinearity in the regression model. Lastly, the leap and bounds algorithm was used to find the variables that explain the models best which you can see on the screen here

Amanda A Webber1:59

Our model contained 18 variables. For the sake of time, we’re showing the three most significant variables as well as the variables that have no significance. The three most significant variables are access to mammograms, access to primary care doctors, and percent of population that is uninsured. The variables that are not significant based on their P value are textile manufacturing, commute, radiance, and unemployment.

You can see from this map of our county residuals that predictive accuracy varied across counties and appears spatially correlated. The R square value reflects the amount of variation in the data set that is explained by our model. On a scale from 0 to 1, this models R Square explains roughly 50% of the variation within the data.

Kai Barner2:47

Big picture, this project wanted to account for spatial heterogeneity. A primary limitation of OLS regression is that it assumes that each explanatory variable has the same relationship with the dependent variable across the entire study area. So when one unit change of an explanatory variable occurs, that results in the same change to the dependent variable, no matter the location. Multiscale Geographically Weighted Regression – MGWR – takes location into account by taking the concept of neighborhoods from standard Geographically Weighted Regression and allowing their size, or bandwidth, to vary between different variables in the data set. Multiple scales reflect diversity in distribution and extent of influence.

This project examined data at the county level, so when analyzing the influence of a variable on an individual county, the bandwidth took into account that many neighbors at the same time. There are just over 3100 counties in the continental US, and we can see that some of the neighborhoods actually encompassed all of them. But at the other end of the spectrum, a small bandwidth indicates that the spatial process changes quickly from location to neighboring locations. And we see this with food index, exercise, access to primary health care, and mammograms.

Where our OLS model had an adjusted R-squared of 0.478, our MGWR model explained about 44 percentage points more variance, with an adjusted R-squared of 0.918.

Here we have our MGWR results for mammograms. The mean coefficient is negative 0.14, so the standard behavior nationwide is an inverse relationship. As the number of mammograms increase, breast cancer mortality decreases and we see this norm in cornflower blue. As we move through the purples into magenta, this negative correlation becomes even stronger. But on the other hand, areas in the cyan extend from having a weaker negative correlation to actually having a positive coefficient. In these areas as mammograms increase, breast cancer mortality also increases.

These local differences warrant further examination of other variables and other factors of underlying influence, alluding to the local complexities behind trying to improve breast cancer health outcomes. There is no one-size-fits-all solution.

Alexandra Diaz Merida5:05

Another variable that we looked at was access to exercise, which measures how close an individual is to a location for physical activity. This can be either parks or recreation centers in this map we see that in general there is a negative correlation, which means that the higher the access to exercise opportunities, there is a decrease in breast cancer mortality. Looking at these graphs, we can see which counties need more access to exercise in order to decrease breast cancer mortality.

Similar to access to exercise, we also looked at food index, which is described as how close an individual is to healthy foods and the and the ability to access due to cost barriers. It is also negatively correlated, which again means that the higher the food index, there’s a decrease in mortality. These maps will help inform local governments about breast cancer mortality rates and how improving these variables can decrease mortality.

In summary, MGWR explains the results better than OLS for spatial heterogeneity, providing better explanations for how urban built environment variables affect breast cancer mortality. Thank you.

Kai Barner6:12

Thank you.

Abigail G Kokkinakis6:13

Thank you.

College of Education and Human Development College of Public Health College of Science Honors College Summer Team Impact Project Undergraduate Research Scholars Program (URSP) - OSCAR

How do sleep, physical activity, and diet contribute to aerobic fitness in U.S. firefighters?

Author(s): Hana Khan

Mentor(s): Joel Martin, Marcie Fyock-Martin; Kinesiology

Abstract: Firefighting involves high intensity work and requires firefighters to maintain optimal levels of health. Unfortunately, due to the nature of their occupation, firefighters’ sleep, diet, and physical activity (PA) are often compromised, contributing to low aerobic fitness.

Purpose: To determine how sleep, PA, and diet behaviors contribute to low respiratory fitness in U.S firefighters. Methods: Participants included [10 males, 3 females, age: 36(30, 41), years of service: 14(10, 18), height(cm): 181.9(177.5, 185), mass(kg): (122.4(109.6, 121.7)] and needed to be Prince William County Firefighters and have no prior history of CVE, pulmonary, renal, and metabolic disease. Participants took a diet (REAP-S), sleep (PSQI), and PA survey (IPAQ-SF). Anthropometric measurements were taken using a BOD-POD®. The 3-minute step test was used to determine aerobic fitness. A Mann-Whitney U test with a p value of 0.05 was used to determine significant variables between two groups: those who performed poorly on the step test versus those who performed above poor. Effect size was calculated using Glass rank biserial correlation coefficient. The project was exploratory, so alpha was set to <0.01. Results- Variable(W-score, p-value): BF %(15, 1), PSQI(18.5, 0.6086), REAPS(18.5, 0.6066), and MPA(21, 0.3458), LPA (18, 0.6709), and Sitting time(12, 0.6709) FFM(11, 0.5541) showed no significant differences between groups. However, FM(4, 0.07593) and VPA(29, 0.02139) were statistically different between the groups. Conclusion: Post- statistical analysis showed that only VPA(29, 0.02139) and FM(4, 0.07593) were statistically different between the two groups. Vigorous physical activity is necessary in firefighters given the intense nature of their work. Excess fat mass is also a hindrance to performance and health as it can increase risk for cardiovascular events. Fire stations need to work towards implementing fitness programs to sustain firefighter aerobic fitness.[/expand] [expand title="Audio Transcript"]Hello! My name is Hana Khan. I'm currently an undergraduate student at George Mason and I am majoring in Community Health with a concentration in Nutrition. I'm so excited to share my research question with everyone that I worked on during my Summer Team Impact Project. So, my research question is: How do sleep, physical activity, and diet contribute to aerobic fitness in U.S. firefighters? You might be wondering, why is this an important or relevant question to ask? As I spent time in the project, it became apparent to me that many firefighters are not able to maintain healthy fitness levels, despite often having to perform physically difficult tasks. Their sleep, diet, and physical fitness levels have been shown in numerous studies to be sub-optimal, and studies have pointed to correlations between these three. Research has shown that aerobic fitness is especially important in firefighters- and yet, many do not have the required fitness levels needed to safely perform their jobs. It turns out, CVD is a leading cause of death in firefighters, most likely due to their work duties. Aerobic capacity is also correlated to CVD, where a low capacity means a higher risk. So I really wanted to focus on that with this topic. There were 13 participants in total, and their specific demographic information is listed on the slide under the methods section. The participants took a few surveys to assess their diet, sleep, and physical activity status. They then had some measurements taken, such as height, weight, body fat, etc. And finally, they performed a three minute step test, which is essentially a test that is used to determine aerobic fitness that involved stepping on and off plates raised to 12 inches to a metronome set to 96bpm. We divided the participants based on how they performed on the 3 minute step test. So one category was for those who performed poor, and one was for those who performed above poor. So after looking at all these variables from the surveys and measurements as seen in the table, and running them through a Mann-U-Whitney test, we found that only Vigorous Physical Activity (which we got from the physical activity survey) and Fat Mass (which we got from the anthropometric measurements) were significantly different between these two groups. Now, I will say that a limitation of this study is definitely that the surveys have self reported values, which means the participants may over or underestimate their responses. However, it cannot be ignored that fat mass and vigorous physical activity are both very important to firefighter fitness and performance. Vigorous physical activity is needed on the job when climbing and carrying things is needed, and excess fat mass can be a hindrance to performance and also increase CVD risk. Thus, I believe it would be beneficial for fire stations to provide individualized fitness programs aimed at targeting aerobic fitness and keeping firefighters active instead of launching them into a physically intense situation from a very low active situation. That's all from me, thank you for listening and I hope you gained something valuable from this presentation![/expand]