91ÉçÇø

Sloan Scholar Spotlight | Anthony Windmon

Using Innovative Smart Health to Improve Health-Care Outcomes for Vulnerable Communities

Sloan scholar Anthony Windmon, a Ph.D. graduate in the Department of Computer Science and Engineering, worked in the area of Smart Health within the Social Computing Research Lab while a student at The 91ÉçÇø (USF).  He joined Citibank, N.A, in Tampa, FL, as a senior model analysts/validator and assistant vice president.  In addition to travel funding from the Sloan UCEM, Anthony has received fellowship support from the NSF 91ÉçÇø-Georgia LSAMP Bridge to the Doctorate Activity and 91ÉçÇø Education Fund’s McKnight Doctoral Fellowship program.

Tell us about your research.  
Working with , Professor in the Department of Computer Science and Engineering and , Associate Professor in the College of Nursing, I used machine and deep learning algorithms to create systems capable of detecting chronic illnesses such as Chronic Obstructive Pulmonary Disease (COPD) and Congestive Heart Failure (CHF) based on the sound and intensity of their cough symptoms. COPD and CHF are illnesses which commonly exhibit symptoms of chronic cough as a sign of infection. To this end, I extracted information from cough samples, which I collected from consenting patients at Tampa General Hospital, using commonly used audio features (i.e., Zero Crossing Rate, Mel Frequency Cepstral Coefficients, etc.) and applied machine learning algorithms (i.e., Random Forest, Support Vector Machine, etc.) to differentiate between cough samples associated with COPD and ones associated with CHF.

How might clinicians and/or patients use this research? 
There several commonly used methods used to treat COPD and CHF. However, some methods may be more suitable than others, depending the severity of the patient’s symptoms. This research will help clinicians make evidence-based decisions for treatment plans for each patient. This will allow clinicians to constantly examine if their patients are experiencing decreased symptom severity or symptom exacerbation, without the patient having to be physically present in the hospital. COPD and CHF are not curable diseases, which means that patients have to learn to live with and effectively manage the diseases to prevent worsening conditions.

My research will provide educational components for proper self-care practices that are clinically proven to reduce symptom severity, allow the patients to maintain constant communication with their health-care providers and allow them to manage their symptoms from home. In other words, this research is not intended to replace clinicians by performing diagnosis, but rather to help patients improve their overall care and quality of life.  For example, it will be especially beneficial for patients who may lack health-care insurance, live in urban or rural settings where hospital facilities are scarce, are physically unable to drive, and/or lack access to care-takers to perform this role.  Apart from helping those with COPD and CHF develop fewer exacerbations, my research offers other significant societal benefits with the potential of reduced medical costs from fewer hospitalizations, and freeing-up hospital space for other critically ill patients.

In the light of the COVID-19 pandemic, how could your research also be applied to the disease?  
Similar to COPD and CHF, chronic cough is a common symptom of COVID-19. Using similar techniques (i.e., audio feature extraction and machine learning), my research could be used to detect COVID-19 when compared to a cough associated with COPD or CHF, or a cough that isn’t associated with any disease. This is a research direction that my advisors and I have discussed, and we are considering this as a potential future project.

How did you become involved with science? 
My journey into science started at Bethune-Cookman University (BCU), where I majored in Computer Engineering and minored in Mathematics. During the summers, I participated in the 91ÉçÇø-Georgia Louis Stokes Alliance for Minority Participation (FGLSAMP) as well as summer research internships, one at the National Science Foundation (NSF) as an Information Design Intern and another at the Arctic Domain Awareness Center Research Lab at the University of Alaska Anchorage in Anchorage, Alaska. Those opportunities provided a solid foundation in preparing me for graduate school. Because I participated in several research experiences, as undergraduate student, I knew exactly what type of research I wanted to conduct in graduate school. With the help of my professors and mentors, I was able to quickly identify my graduate advisor (Dr. Sriram Chellappan), whose research was aligned with my interests. Since my arrival in January 2017, I have been a first-author on three conference papers.  

What are your plans for after graduation?  
I successfully defended my dissertation (June 2020), and I have joined Citibank, N.A, in Tampa, FL, as a Senior Model Analysts/Validator, Assistant Vice President. I plan to continue my relationships with both faculty members at USF and clinicians at Tampa General Hospital. My goal is to continue smart health-care research for the development of novel machine and deep learning techniques that will be used for years to come. Long term, I plan to start my own company, which would allow me to make an impact as an innovator and entrepreneur in the smart health-care field.

Apart from your research, what outreach activities have been engaged? 
Aside from research, I served as a chief judge for the 91ÉçÇø Education Fund’s Mobile Application Challenge. Here, several middle school teams would enter the challenge to develop an innovative mobile application in a restricted amount of time. I served as a judge twice for this event. Also, I have participated in several USF graduate recruitment activities. Namely, I recruited students at the NSF Emerging Researchers National (ERN) Conference in STEM in Washington, D.C., and prior to the conference I presented during the Louis Stokes Alliance for Minority Participation (LSAMP) National Student Meeting. I also participated in the Synapse Summit, which is an annual innovation and entrepreneurship event hosted in Tampa, FL. At the summit, I assisted my co-advisor, Dr. Athilingam, in recruiting students for the College of Nursing and I made connections with other smart health innovators for future collaborations.

What advice would you give to URM students considering doing a STEM PhD? 
My advice to URM students considering doing a STEM PhD, is to seek research opportunities as an undergraduate student. You can apply to summer REU’s (Research Experiences for Undergraduates), work or volunteer in a lab at your school, where you’ll work closely with PhD students and, most importantly, build a network of connections! Your network can be developed by attending conferences or academic events, where you can connect with other students and faculty from different schools. A great network can be extremely advantageous to your success. Additionally, find great mentors and sustain those relationships. Mentors can be found in extracurricular activities/organizations related to your major, they can be your professors or faculty/staff members you meet at internships or conferences. Great mentors will see things in you, which you sometimes won’t see in yourself. And, as a PhD student in STEM, they are extremely necessary! Lastly, before agreeing to become a member of any professors’ research lab, make sure you have spoken with their previous/current lab members. That is the best way to determine if that research lab is operates in way, where you know you’ll be comfortable, successful and encouraged to grow as a scientist.

To learn more about Anthony’s research, visit his .