The 5th Annual Data Sciences Hackathon was held virtually over the course of two days on September 9th and 10th, hosted by the UAB Department of Biomedical Informatics and Data Science (DBIDS) in conjunction with the Center for Clinical and Translational Sciences (CCTS). External sponsors for the event included the Pittsburgh Regional Health Initiative and the Jewish Healthcare Foundation.
This year’s theme was "Intelligent Safety: Pioneering Patient Safety Solutions with AI/ML [Artificial Intelligence/Machine Learning] and Data Science," which placed a critical focus on how items such as natural language processing (NLP), generative AI such as ChatGPT, and computational resources impact nearly every aspect of patient care.
One of the primary goals for Hackathon was to strengthen the professional development of participants through training, mentoring, and networking, while encouraging the development of novel research questions and hypotheses for future work in addressing patient safety. In preparation for this year’s Hackathon, there were several workshop tutorials held that centered around the event theme, while focusing extensively on generative AI as well as natural language processing.
For the duration of Hackathon, participants were able to ask questions to their mentors in reference to the projects being worked on, with most of their work going into the early hours of the morning. Amy Wang, M.D, MBI, Associate Professor of Medicine and chair of this year’s Hackathon, expressed that the opportunity for collaboration in an events such as this could only serve as a benefit in the long run.
“It is my duty and my pleasure to help our Hackathon community network, learn about important problems in biomedicine, learn about resources and methods for solving these problems, and develop collaborative, technical, and problem-solving skills,” said Dr. Wang. “I am always amazed at how teams are able to gather, discuss, design, prototype, implement, and test innovative solutions in only two days. I am grateful for the opportunity to bring everyone together to build research communities and especially grateful for the participants, sponsors, and volunteers who make it all possible.”
At the conclusion of Hackathon, teams were able to present their projects at the Innovation Showcase the following week, where judges evaluated their projects using criteria similar to those used for scientific publications and grant proposals, such as significance, approach, and innovation.
“With the Hackathon, it is very rewarding to see how much you can accomplish in such a short period of time, and I believe that because there is such a small-time frame, there is really no pressure to be perfect,” said Data Information Coordinator Melissa Hall. “A lot of times in academia, we don't want to show our work with the public until it's fully formed. The hackathon is very refreshing because no one has anything close to being published. Everyone is very encouraging, and it is such a positive atmosphere.”
A common takeaway by participants from this year’s Hackathon was the opportunity to work in a space where imagination and originality was not only allowed, but also encouraged.
“[Hackathon] was able to provide a platform to apply academic knowledge to real-world problems, fostering practical skills and experience,” said Phuong Quach, MPH and Ph.D. candidate in Rehabilitation Sciences for the UAB School of Health Professions. “It encourages interdisciplinary collaboration, allowing participants to learn from experts in various fields, while promoting innovation, creativity, and out-of-the-box thinking, which can lead to groundbreaking solutions.”
After the Innovation Showcase, winning teams were given awards for the projects along with cash prizes. For their efforts, the first prize-winning team received $2,000, the second-place team received $1,500, and the third-place team received $1,000. The winning teams, projects and team members were as follows:
First Place Team: DeepSepsis
Title of Project: PhLORENS: An AI-based Application for Early Detection of Sepsis in the ICU
Team Members: Kevin Song, (Ph.D. Candidate in Biomedical Engineering), UAB, Zhandos Sembay (MS, Data Analyst, UAB Informatics Institute), Melissa Hall, MS (Data Information Coordinator, UAB Informatics Institute) and Radomir Slominski, Ph.D. (Postdoctoral Fellow, UAB Heersink School of Medicine)
Second Place Team: SSO: Sepsis Special Ops
Title of Project: k-Means Unsupervised Clustering for Subtyping Patients with Sepsis
Team Members: Anna M. Chang (Associate Faculty in Biology, Edmonds College and DO Candidate), Pacific Northwest University of Health Sciences, Van Huynh, MS, (Ph.D. Candidate in Biomedical Sciences), UAB Heersink School of Medicine, Yishu Qu, MS, MPH (Ph.D. Candidate in Health and Biomedical Informatics (HBMI)), Northwestern University Feinberg School of Medicine, Phuong Quach, MPH (Ph.D. Candidate in Rehabilitation Sciences), UAB School of Health Professions
Third Place Team: Hold the Line
Title of Project: Exploring Central Venous Catheter Infections in UAB EHR Data
Team Members: Curtis Hendrickson, Bioinformatician, Department of Microbiology (UAB Heersink School of Medicine), Nate Hendrickson, UAB Dual-Enrolled High School Student
With AI now more prominent than ever, Dr. Wang expressed how important it is for future Hackathons to reflect the current status of the world.
“AI will become prominent throughout everyday life, including work, patient care, and research,” said Dr. Wang. “As Hackathons are focused on data, teams will use new, adapted, and combined AI methods to develop solutions that will contribute to discussions, coding, and testing.”
Although participants of this year’s Hackathon project different individual impacts from being in Hackathon, their contribution to helping the Hackathon community network grow is immeasurable.
“Collaborating with individuals from diverse backgrounds who brought unique skills and perspectives to the table was enlightening, and it has motivated us to keep looking for novel solutions in the field of healthcare AI,” Phuong Quach said. “The Hackathon offered a unique opportunity to apply our skills directly to real healthcare problems, enhancing our understanding of their practical applications in the field.”
“Our Hackathon community network can only be helped with events such as these in learning about important problems in biomedicine, learning about resources and methods for solving these problems, and developing collaborative, technical, and problem-solving skills,” said Dr. Wang.
Hackathon 2023 proved to be another vital step in research groups improving their abilities to gather, design, test, and implement innovative solutions, while building research communities that look to improve the overall status of health care treatment.