Please save the date! We’re looking forward to holding our next Duke Health Data Science Poster Showcase on Friday, December 6, 2024. All are welcome, and a light lunch and refreshments will be served.
Save the date for the third annual poster showcase Read More »
Duke University and Duke-NUS Medical School in Singapore’s AI + Medical Sciences Initiative (DAISI) invite you to a virtual symposium on AI, in partnership with the Duke AI Health Community of Practice.
This two-hour event on Thursday, June 6 from 8am – 10am (EDT) will feature lightning talks — short, fast-paced presentations — presented by faculty and staff involved in health-related topics.
Virtual Symposium on AI Presented by DAISI (Duke NUS AI + Medical Sciences Initiative) Read More »
By Jessica Johnstone
A successful poster session held at the Duke’s School of Nursing this past March marked the completion of a new writing workshop series conducted by the Duke AI Health Community of Practice. This initiative, titled “Foundations of Scientific Writing for Staff Members,” was focused on helping operational staff develop and extend their scientific and scholarly communication skills.
“Although we made use of a traditional lecture-style format, we also designed the series with an eye to interactivity and student participation,” said Jonathan McCall, MS, communications director for Duke AI Health.
McCall, who helped lead the didactic portions of the workshops, went on to note that the organizers didn’t want the virtual classes, held via Zoom, to consist solely of “talking head” instruction.
“One feature that arose almost through improvisation – using small breakout sessions during the classes – helped keep things engaging for our learners,” he added.
The workshop series invited individuals with varying degrees of experience in scientific writing but was particularly focused on encouraging participation from a broad spectrum of staff, including coordinators, project managers, project leaders, and analysts. Learners had the opportunity to choose poster topics that aligned with their interests and experience in operational processes, program descriptions, communication strategies, and more.
“I’m thrilled to witness the growth and enthusiasm among participants as they honed their scientific writing skills throughout the workshop series,” said project manager Jessica Johnstone, who helped organize and run the workshop series while also participating in it as learner.
The workshop series, which was produced in collaboration with the Duke Clinical and Translational Science Institute (CTSI), comprised five sequential sessions that addressed basic principles of scholarly writing and publication and provided a foundation for hands-on experience developing posters. Although classes were conducted remotely, the culminating poster session, held on March 13th at the Christine Siegler Pearson Building at the School of Nursing, was an in-person event. Incorporated into the first day of the Fostering AI/ML Research for Health Equity and Learning Transformation (FAIR HEALTH™) symposium hosted by DUSON professor Michael Cary, PhD, RN, the interactive poster session gave participants a forum to showcase their work and gain experience both in presenting and discussing it.
“Our goal for this workshop series was to create a valuable addition to our employees’ training and development,” said workshop organizer Shelley Rusincovitch, MMCi, FAMIA, managing director of Duke AI Health and co-director of the CTSI Biomedical Informatics & Data Science (BIDS) Pillar. “We’re very glad to see this program succeed and we appreciate the support of CTSI’s leadership in making it possible.”
Input from ChatGPT (chat.openai.com) was used to revise and edit this article.
Scientific Writing Workshop Series Empowers Staff Read More »
We are thrilled to invite you to our upcoming symposium, Fostering AI/ML Research for Health Equity and Learning Transformation (FAIR HEALTH™), scheduled for March 13-14, 2024, at the Duke University School of Nursing in Durham, NC. This two-day event is dedicated to advancing discussions on cutting-edge research and practices aimed at promoting equity and fairness in algorithmic systems.
Attendance is free and open to everyone. Please join us! Register at https://duke.qualtrics.com/jfe/form/SV_6RIrLwyKEyM8FOm
Monday, January 8, 2024: Duke University recently hosted the Fostering AI/ML Research for Health Equity and Learning Transformation (FAIR HEALTH) Workshop at the Kirby Horton Hall in the scenic Sarah P. Duke Gardens in Durham, NC. This inclusive event was designed for individuals passionate about advancing healthcare through innovation while emphasizing equity and fairness in clinical algorithms. Attendees, including faculty, staff, and students, comprised a diverse audience committed to shaping the future of healthcare technology.
The FAIR HEALTH Workshop focused on the pressing issue of algorithmic bias in clinical decision-making, offering participants valuable insights and strategies to identify, evaluate, and mitigate biases in healthcare algorithms. With AI and machine learning playing increasingly pivotal roles, the imperative was to ensure these technologies don’t inadvertently perpetuate biases, promoting unequal treatment.
A distinguished panel of experts, featuring Duke AI Health’s Michael Cary, PhD, RN, Sophia Bessias, MPH, MSA, and Ben Goldstein, PhD, and Duke-Margolis Center for Health Policy’s Christina Silcox, PhD, led discussions into the legal and ethical dimensions surrounding the implementation of clinical algorithms. The panelists offered practical strategies applicable across the entire development lifecycle. Their aim extended beyond raising awareness about the challenges posed by algorithmic bias; they sought to equip attendees with the tools needed to effectively address these challenges.
The workshop’s contributions were pivotal, illuminating a path for advancing healthcare equity through innovative AI and machine learning solutions that enhance patient care while minimizing bias and prioritizing equity.
Another distinctive element of the FAIR HEALTH Workshop was its interactive nature. Organizers seamlessly integrated a combination of lectures and an engaging case study discussion to foster active participation. The case study honed in on a clinical prediction algorithm deployed at Duke Health, allowing participants to apply their bias-probing skills in a real-world scenario.
The panelists underscored the importance of considering legal and ethical implications in the development and deployment of clinical algorithms. The complex regulations surrounding AI in healthcare were demystified, providing clear guidance to participants. This knowledge is pivotal in navigating the evolving landscape of healthcare technology, ensuring that ethical considerations stand at the forefront of AI integration.
Attendee feedback underscored the success of the workshop, receiving a stellar rating of over 4.5 out of 5. The interactive discussions and practical examples emerged as key components that significantly enhanced the understanding of how biases can infiltrate clinical AI tools. Participants expressed gratitude for the opportunity to hone their skills in assessing data reliability and reducing bias, crucial elements for preventing AI systems from inadvertently favoring certain patient groups.
The FAIR HEALTH Workshop at Duke University marked a substantial stride in ongoing efforts to ensure fairness in AI used in patient care. The event successfully brought together a diverse group of professionals, fostering collaboration across disciplines, including clinicians, clinician scientists, social scientists, and technical experts. As AI integration in healthcare continues to burgeon, the commitment to its ethical application extends beyond professional boundaries, becoming a community responsibility.
By Jessica Johnstone
Friday, December 8, 2023 was a lively day as the Sarah P. Duke Gardens, Kirby Horton Hall was transformed into a hub of quantitative activity, hosting the Health Data Science Poster Showcase. It included 55 posters, each featuring insightful research, showcasing the depth and talent of Duke’s diverse community. From statistics and informatics to machine learning and data engineering, the topics spanned the full spectrum of health data science. Students, fellows, staff, and faculty alike presented their work, illuminating the collaborative spirit that thrives at Duke. The event, championed by Duke AI Health, the Duke Clinical & Translational Science Institute, the Laboratory for Transformative Administration, and the Center for Computational Thinking, was a testament to Duke’s commitment to fostering innovative research.
The event was a resounding success, fostering connections, sparking ideas, and leaving everyone energized by the potential of data-driven health. “The showcase illuminated the depth and impact of health data science across Duke,” Shelley Rusincovitch, MMCi, the event’s director said. “Empowering these collaborations is precisely why we do what we do.”
Michael Pencina, PhD, Chief Data Scientist for Duke Health and Director of Duke AI Health, extended a warm welcome to the audience. He remarked, “This endeavor reflects Duke’s dedication to ethical equitable data science, showcasing the application of advanced natural language processing methods to extract meaningful insights for the future of healthcare.”
Akshay Bareja, PhD, an Assistant Professor in Duke Molecular Physiology Institute, presented the Best Computational Thinking awards from Duke’s Center for Computational Thinking, to Tanner J. Zachem, a PhD student in the Department of Neurosurgery at the Brain Tool Laboratory under Patrick Codd, and Aditya Parekh, MS, a Duke AI Health Data Science fellow in Rohit Singh’s lab. Dr. Bareja commented, “There were so many excellent posters presented. I was particularly struck by the novelty and sophistication of the winner (TumorID) and runner-up (Sceodesic). The Tumor ID team developed a florescence spectroscopy device that can accurately classify tumor type and shows promise as a tool that can be used during surgery to perform real-time tumor identification and classification. Inspired by how flight paths are computed, the Sceodesic team developed a novel gene-program discovery algorithm. The clear and immediate real-world impact of TumorID gave it the edge and the fact that the presenter, Tanner Zachem, is still a first-year PhD student made the achievement all the more impressive!”
Mr. Zavhem’s poster, “Machine Learning and Laser Induced Fluorescence Spectroscopy for In Vivo Brain Tumor Identification” included co-authors Rory Goodwin MD, PhD, and Patrick J. Codd MD. Mr. Parekh’s poster “Sceodesic: Navigating the Manifold of Single-cell Gene Coexpression to Discover Interpretable Gene Programs” included co-authors Sinan Ozbay, MFin, and Rohit Singh, PhD.
Duke’s Quantitative Expertise Shines at the Health Data Science Poster Showcase Read More »
Duke Health Data Science Poster Showcase
Happening tomorrow: Friday, December 8, 2023 | 11:00 AM – 1:00 PM (Eastern time)
In person at the Sarah Duke Gardens, Kirby Horton Hall
The fall 2023 Health Data Science Poster Showcase will highlight the transformative work in health data science happening all across Duke. We’re excited that the showcase will feature 54 posters – our largest-ever showcase – with participation by departments, centers, and institutes from all across Duke.
We will serve a light lunch from 11:00-1:00, and we invite you to join the group photo at 12:00 PM and the poster award presentations at 12:30 PM. You can drop in for a few minutes, or stay for the entire time!
All are welcome, and we invite you to join us! Please do feel free to share and forward to others.
The event will be held in the Sarah Duke Gardens, in the visitor center Kirby Horton Hall (420 Anderson Street, Durham, NC 27708). This is just a short walk from the main hospital and clinics, and close to the C1 and H1 Duke Transit Routes. Free, reserved parking for this event is next to the building (through the Anderson Street entrance).
Please join us for the next Health Data Science Poster Showcase Read More »
A groundbreaking paper co-authored by Duke Health’s Nicoleta Economou, PhD, and Michael Pencina, PhD, recently published in NPJ Digital Medicine discusses leveraging a Quality Management System (QMS) for AI/ML development intended for healthcare. The authors explain how a tailored QMS framework can bridge the AI translation gap, ensuring safe, ethical, and effective incorporation into patient care. This approach can accelerate the translation of AI research into practical clinical applications, prioritizing patient safety and fostering trust in healthcare innovation.
FAIR HEALTH (Fostering AI/ML Research for Health Equity and Learning Transformation)
Monday, January 8, 2024, 9:00 AM – 12:00 PM
In person at Sarah Duke Gardens in Kirby Horton Hall
This event is open to anyone who is passionate about advancing healthcare through innovation while ensuring health equity and fairness in clinical algorithms, including faculty, staff, and students.
This workshop will delve into the critical issue of algorithmic bias and clinical decision making. This is an opportunity to gain essential insights and practical strategies to identify, mitigate, and evaluate bias in clinical algorithms. We will also explore the legal and ethical implications of algorithmic bias in healthcare, an aspect that is gaining increasing importance in today’s dynamic healthcare landscape. To enrich the workshop and encourage active participation, we have incorporated a combination of lectures and an interactive case study discussion, making it an even more rewarding experience for all attendees. By engaging in this workshop, and learning from one another, we can pave the way for a future where healthcare algorithms enhance patient care, minimize bias, and prioritize equity.
Registration open for FAIR HEALTH workshop on January 8, 2024 Read More »
Vanderbilt University Medical Center (VUMC) and Duke University School of Medicine were awarded a $1.25 million grant from the Gordon and Betty Moore Foundation for the project “Measuring Artificial Intelligence (AI) Maturity in Healthcare Organizations.” Working with the Coalition for Health AI (CHAI) and the University of Iowa, a team of experts will leverage the grant to develop a maturity model framework. The project leads are Peter Embí, MD, MS, and Laurie Novak, PhD, MHSA, from VUMC; and Michael Pencina, PhD, and Nicoleta Economou, PhD, from Duke. This framework will outline the essential capabilities that health systems must establish to ensure they are well-prepared for the trustworthy utilization of AI models.
|
