Over the summer, a UTSW team called Phixer began developing a product using artificial intelligence (AI) to reduce physicians’ documentation burden, improve diagnostic accuracy, and improve patient outcomes .
Phixer’s efforts revealed the potential of AI technology to advance healthcare, winning first place in UT Southwestern’s inaugural AI Innovations Challenge, open to UTSW employees, learners and interns. The group used AI-based conversation analysis and diagnostic support to summarize doctor-patient interactions and provide real-time insights. The team was among three to present innovative AI-based projects.
The audience at the pitch event, which includes team members, judges, industry partners and mentors, applauds the innovative AI proposals.
The goal of the challenge was to foster interdisciplinary collaboration among UTSW clinicians, researchers, students, interns and scientists to develop practical applications of AI in life sciences, biotechnology and healthcare.
Co-sponsored by the Technology Development Office (OTD) and the Lyda Hill Department of Bioinformaticsthe challenge was open to UTSW staff. The event focused more on collective learning and development than direct competition, organizers said.
“At UT Southwestern, there is a desire to integrate cutting-edge technology into everyday health care as a leading academic medical center. Healthcare, research and medical education would benefit immensely from the integration of artificial intelligence,” said Prapti Mody, Ph.D., program manager in the Lyda Hill Department of Bioinformatics.
In June, three teams consisting of 12 participants in total were selected for the program. From July 1 to August 28, team members attended lectures from mentors and partners working in healthcare and educational institutions across the country to help them gain skills crucial marketing skills through virtual and live courses on financial literacy, legal and policy issues, regulation. business, financing and other topics.
Simultaneously, they began designing and testing their AI projects with important guidance from UTSW mentors and industry partners. The mentors included Hayden Blackburn, former COO of TechFW, who provided two coaching sessions to each team to help them prepare their pitches. Eric Zimmerman, director of healthcare and life sciences business development, venture capital and startups at Amazon Web Services, provided Amazon Web Services credits to help participants develop their AI tools.
The challenge concluded on August 29 at the AW Harris Faculty Club when teams presented their ideas to an audience comprised of industry partners and UTSW clinicians, academics and staff. At each table, audience members vote for the first, second and third place winners. Proposals were judged on their uniqueness, relevance to healthcare issues and commercial viability.
Physical therapist Abigail Offringa, PT, DPT, a member of the Phixer team, explains her group’s goal of using AI to help doctors reduce their documentation burden. Phixer won first place in the challenge.
Since most of the Phixer team members are physical or occupational therapists with clinical training, they viewed the challenge as a new, exciting and invigorating learning experience, said Hari Vennelakanti, PT, DPT, chief team member and certified physiotherapist in lymphedema oncology.
“It wasn’t just about the outcome; it was a collective effort, a shared commitment and the belief that we could achieve something remarkable together,” said Dr Vennelakanti. “We dove into the world of business and technology, thinking and building like a startup. The competition provided us with the tools needed to successfully integrate AI into healthcare.
UMed Imaging, which developed a model to minimize patient radiation exposure by converting medical images into multiple modalities, took second place. MedBox, which offered an automated grading and feedback assistant for medical training programs and simulation centers, came in third.
Streamlining primary care
Phixer focused on developing a tool that eases the documentation burden for clinicians and improves the quality of patient care. The team used machine learning (ML) technologies to transcribe doctor-patient conversations in real time, offer diagnostic assistance, and generate comprehensive visit summaries, said Nicole Wiggs, PT, DPT, chief of Phixer team, physiotherapist and head of outpatient therapy.
“The overall goal was to reduce the administrative burden on providers and devote more time and space to human interaction,” Dr. Wiggs said. “Connecting, listening and helping patients are why we all came to the medical field, and AI will be a great solution to achieve this.” »
Phixer team members include, from left: Sean Talluri, MS, MBA, Rhoda Talluri, PT, D.Sc.PT, Abigail Offringa, PT, DPT, Lina Asfoor, OTR, OTD, CLT, Nicole Wiggs , DPT, PT and Hari Vennelakanti, PT, DPT
Since its victory, Phixer has worked to refine its prototype and gather feedback from primary care providers and other stakeholders.
“As we move forward, we are excited about the potential of our product to revolutionize the way documentation is managed in healthcare settings,” said Dr. Vennelakanti. “We believe that with the right support and continued dedication, we can bring this innovative solution to market and make a significant difference in the healthcare industry.” »
Decrease in radiation exposure
UMed Imaging has developed a deep learning model capable of converting between multiple medical imaging modalities, such as CT scans, PET scans and MRI scans.
“This innovation has the potential to streamline radiotherapy processes by reducing the need for multiple examinations,” said Yunxiang Li, team leader and student at the university. Biomedical Engineering Graduate Program. “It also reduces health care costs and minimizes radiation exposure.”
UMed Imaging team members include, from left: Yunxiang Li and Jiacheng Xie.
In radiotherapy, multiple imaging modalities are used to optimize treatment plans and evaluate therapeutic effectiveness, Li said. The lack of an effective universal modality conversion model hinders the integration of these different methods, because it is usually technically, logistically, and financially difficult to acquire them all for each patient.
“Our model will help reduce clinical burden, decrease radiation doses, and reduce medical costs and time, which in turn will optimize the entire radiotherapy process, providing patients with safer and more effective treatments. effective,” he said.
Automation of medical education
MedBox, the team of three people from Jamieson Laboratory in Lyda Hill’s Department of Bioinformatics, focused on using AI to automate grading and feedback for medical education programs used in the UTSW Simulation Center. The project was developed in collaboration with the Simulation Center.
MedBox team members include, from left: Dhavi “Anni” Jain and David Hein, MS. Not pictured: Ameer Hamza Shakur, Ph.D.
Typically, a Sim Center medical student conducts a simulated visit with a patient-actor and writes an encounter note after the visit, said team leader David Hein, MS, Data Scientist. These notes are then noted by hand.
MedBox includes a custom hardware and software “box,” tailored to institutional needs, of pre-organized pipelines deploying large language models. This strategy allows MedBox to manage time-consuming tasks.
One example involves rating videos of medical students’ encounters with patient-actors.
“We have tools that create a transcript from the video recordings and then use that transcript as text input into a model to detect whether students performed the correct physical exam,” Hein explained.
The patient-actor may have allergy symptoms, for example, and part of the student’s grade for this practical scenario is to perform an ear exam. Automated scoring focuses on technical details and allows patient stakeholders to prioritize the assessment of the student’s communication and interpersonal skills.
“This approach would improve health care by allowing medical students to practice more patient encounters,” Hein said. “If scoring and feedback from AI systems could be faster, students could have many more patient encounters – and it would free up instructors, too. »
These exciting ideas are just a glimpse of what the future holds for AI possibilities, said Brad Phelan, MBA, associate vice president of technology commercialization and business development.
“A key facet of OTD’s mission is to advance UT Southwestern technologies into the commercial market to positively impact patients’ lives, and the AI Innovations Challenge is a great example of a program supported by OTD to help us achieve this mission,” said Mr. Phelan.
