Biomedical Informatics and Medical Education > News and Events

News and Events

Chair’s Message

pth-use-this-one We are moving toward our vision with a number of activities across our various programs. We have updated our strategic plan in response to the 10-year academic program review that we recently completed. For our research-oriented MS and PhD programs, we have recently added a specialization in Data Science. We are completing a curriculum revision for our on line applied clinical informatics MS which will be effective Fall 2020. The work of our fellows in the clinical informatics fellowship program has received plaudits from clinical administrators and faculty, and we are currently recruiting a new faculty member in our department to assist with this program (view position description). We are also recruiting a faculty member in medical education to start Summer 2020 (view position description). This is the beginning of a new cycle of admissions to our graduate programs, and we look forward to another productive year, and new growth in our department.

Cordially,

Peter Tarczy-Hornoch, MD
Chair and Professor, Department of Biomedical Informatics and Medical Education

Biomedical Informatics and Medical Education Newsletter

April 28, 2025 – May 2, 2025

UPCOMING LECTURES AND SEMINARS
BIME 590
Presenter: Dr. Thinh Tran, PhD
Thursday, May 8^th – 11-11:50 am
850 Republican Street, Building C, Room 123 A/B
Zoom Information: https://washington.zoom.us/my/bime590
Speaker will present via Zoom

Title:
Assessing performance of large language models in clinical data abstraction and reasoning from unstructured clinical text

Abstract:
Patient health records contain a wide array of clinical information of high value. However, this information often resides in unstructured notes, making systematic computational analysis challenging and posing a major bottleneck for precision medicine. Recent advancements in knowledge extraction and reasoning have positioned large language models (LLMs) as a promising automated approach for clinical information extraction and inference. Here, we systematically evaluate the effectiveness of various LLMs by building a flexible and scalable software pipeline. Our pipeline ingests unstructured medical records and extracted specified clinical outputs, compatible with all mainstream commercial and open weight LLMs in a HIPAA-compliant manner. Using a human-annotated clinical note dataset of 843 unstructured medical records, we employed various LLMs for two key tasks: 1) extracting clinical information of cancer medication name and treatment duration; and 2) inferring treatment response based on longitudinal clinical notes and minimal residual disease (MRD) measurements. For information extraction, we evaluated Llama 3.3, Claude 3.5 Sonnet, GPT-4o, and DeepSeek-R1-8B, using an array of prompting strategies. Claude 3.5 achieved the highest accuracy for both medication name (F1=0.72) and treatment duration (median difference of 0 and 37 days for start and end date, respectively), but at a higher cost compared to other LLMs. Notably, most LLMs (except Llama 3.3) performed best with simple prompting that directly states the information to be abstracted, showing minimal gains from additional instructions or examples. For reasoning, we assessed both general-purpose LLMs (Claude 3.5, GPT-4o) and reasoning-focused LLMs (DeepSeek-R1, GPT-o1-mini, GPT-o3-mini) on their ability to infer treatment response by integrating clinical notes, imaging results, and MRD measurements. Unlike in math and programming where reasoning LLMs significantly outperform regular LLMs, both categories of LLMs demonstrated comparable capabilities in identifying and integrating relevant clinical evidence, even though the final conclusions varied by each method. These preliminary results suggested future improvements by reasoning LLMs and test-time scaling in the highly-specific clinical domain. Our results demonstrate the clear operational value of LLMs as a rapid, cost- and time-efficient means to comprehend clinical information for scientific advances and ultimately improved patient care.

Speaker Bio:
Thinh Tran is a Bioinformatics Data Scientist at Natera, where she applies AI/ML methods to clinical and genomic data to advance therapeutic discovery. She completed her PhD in Cancer Biology at Memorial Sloan Kettering Cancer Center, focusing on natural language processing for treatment extraction from clinical notes and on benchmarking variant effect predictors on annotating cancer variants of unknown significance and identifying novel prognostic biomarkers.

BIME 591
Wednesdays – 11:30-12:20 pm
Section B, HSEB 421
Zoom Information: https://washington.zoom.us/my/velvinfu?pwd=Y3dJbjNBeFpzTC9HTXV1UDFYYXlKQT09

ANNOUNCEMENTS
On May 1 2025, Doug Schaad will have completed 50 years of employment with the UW School of Medicine!

_______________________
By Monday, May 5th, please RSVP for BIME’s end-of-year celebration, where we recognize our students, postdocs, fellows, and great teachers and mentors.
Event Details

Friday, June 13^th
2-3 pm: Recognition ceremony in Orin Smith Auditorium or via Zoom
3-5 pm: Reception in C123

Students, postdocs, and fellows are welcome to bring guests. Please indicate the number of in-person guests on the RSVP form.
A Zoom link for the hybrid recognition ceremony will be sent at a later date.
We hope you can join us!

_______________________
Have a new idea that you think will transform medicine? If so, be sure to review the details in our 2025 RFA and submit a proposal by May 23rd, 2025.

We are seeking applications for one-year pilot projects in the area of medical data science, for potentially up to $45K, with $25K guaranteed direct costs + the possibility of adding up to $20K in cloud credits, pending agreement with cloud service providers.

Please note that only UW faculty are eligible to apply. Learn more here.

_______________________
UPCOMING WORKSHOP
Considering Cultural and Linguistic Diversity in AI Applications: A Hybrid Workshop

AI has tremendous potential to improve all different aspects of our lives – leading to changes in many facets of life, including making our day-to-day lives easier, allowing for more personalized healthcare, and generating insights as we live and learn. There are also challenges and potential issues that we should consider, such as:

How are people affected differently by AI?
How might aspects of one’s culture, background, and other characteristics or traits affect how we might respond to AI?
How might these considerations inform how we regard or develop applications involving AI?
How do we collaboratively work towards integrating the values and preferences of communities of interest in AI tools?

This workshop will be a hybrid event on May 6, 2025 from 12:30pm to 16:30pm PST (time zone converter here). It seeks to bring together researchers, practitioners, and others who work with (or are interested in working with) culturally and linguistically diverse (CALD) populations. All levels of technical expertise welcome!

Event purposes:

Raise awareness about the importance of thoughtful engagement in the design of AI applications for culturally and linguistically diverse populations;
Offer a venue to engage in activities to think about relevant concepts and build relevant skills; and
Present and hear about the work of others on this topic.

Workshop locations: University of Washington (UW); University of California, Irvine (UCI); and virtual.
Event Sponsors:

Association for Information Science & Technology (ASIS&T) U.S. West Chapter
UCI Connected Learning Lab

Accepted work including UW BIME folks here: https://cald-ai.github.io/accepted-works/.
More information and how to register here: https://cald-ai.github.io/
Contact Annie T. Chen <atchen@uw.edu> with questions.
_______________________
2025 UW Wellness Symposium
Enhance your well-being at the 2025 UW Wellness Symposium, Wednesday, May 21, and Thursday, May 22. Join expert-led virtual sessions on mental, physical and financial health with The Whole U.

PAPERS & PUBLICATIONS

Oliver Li. “Quantifying Controversy: A Novel Approach to Detecting Misinformation” has been accepted to the conference and proceedings at MEDINFO 2025 in Taipei, 11 Aug 2025.

Chunyan Duan, Shijun Chen, Jiajie Wang, Qianqian Tong, Qiantuo Liu, Faisal Yaseen, Shouyi Wang, Daniel S. Hippe, Stephen R. Bowen .”Two-Stage Clustering and Auto Machine Learning to Predict Chemoradiation Response in Tumor Subregions on FDG PET for La-NSCLC” has been accepted as a SNAP ORAL PRESENTATION in the session “Radiopharmaceuticals, Theranostics, and Nuclear Medicine SNAP Oral: Therapy III” for the 67^th Annual Meeting & Exhibition of the American Association of Physicists in Medicine in Washington, DC, July 27 – 30, 2025.

Chunyan Duan, Han Zhou, Jiajie Wang, Qiantuo Liu, Faisal Yaseen, Shouyi Wang, Daniel S. Hippe, Stephen R. Bowen^.“Muilt-Instance Learning Model with 2D and 3D Features Representation and Transformer-Based Prediction for FDG PET Tumor Chemoradiation Response of La-NSCLC” has been accepted as a General Poster Discussion Poster Presentation for the 67th Annual Meeting & Exhibition of the American Association of Physicists in Medicine in Washington, DC, July 27 – 30, 2025.
Chunyan Duan, Jing Sun, Jiajie Wang, Qiantuo Liu, Xiaojing Zhu, Faisal Yaseen, Shouyi Wang, Daniel S. Hippe, Stephen R. Bowen^.“A Two-Layer, Two-Task Prediction Model Based on 3D Imaging and Residual Networks for Mid-Chemoradiation Tumor Response Prediction on FDG PET for La-NSCLC” has been accepted as a General Poster Discussion Poster Presentation for the 67th Annual Meeting & Exhibition of the American Association of Physicists in Medicine in Washington, DC, July 27 – 30, 2025.

UPCOMING EXAMS
Title: Computational Approaches to Predict Health Outcomes Using Cytometry Data
Student: Ya-Lin, Chen
Date/Time: Tuesday, May 13, 2025, 11 AM – 1 PM (Pacific Time)
Location: 850 Republican Street, Building C, Room 359
Zoom: https://washington.zoom.us/j/92178550574

Abstract: Flow cytometry plays a pivotal role in medicine, as demonstrated by the widespread use of the complete blood count (CBC) across medicine. Alongside summary markers such as cell counts, CBCs also generate rich single-cell datasets, which are drastically underused. To allow for enhanced clinical use of these single-cell data streams, there is a need for methods which can generate clinically relevant biomarkers, or which can directly use these data for clinical prediction. To tackle these issues, we propose three primary research aims: (1) algorithmically developing novel markers from clinical single-cell data using unsupervised techniques, (2) evaluate and develop novel single-cell data modeling methods for predicting health outcomes, and (3) applying these single-cell approaches to significant health conditions, including predicting pregnancy and cardiac surgery complications and predicting treatment efficacy for hematologic malignancies. Approaches will be validated using large-scale data archives from University of Washington Medical Center (UWMC), alongside multiple open-source single-cell datasets. Collectively, this project will lead to generation of novel computational methods which can create tangible benefit for multiple well-defined clinical challenges.

April 21, 2025 – April 25, 2025

UPCOMING LECTURES AND SEMINARS
BIME 590
Presenter: Dr. Halil Kalicoglu, PhD
Thursday, May 1^st – 11-11:50 am
Zoom Information: https://washington.zoom.us/my/bime590
Speaker will present via Zoom Only

Title:
Enhancing Rigor and Integrity of Biomedical Research using Natural Language Processing

Abstract:
There has been much debate about rigor, transparency, and reproducibility of biomedical research in recent years. Ongoing efforts aim to address issues in research conduct and reporting by developing standards, guidelines, and recommendations. As biomedical research output increases exponentially, automated tools are needed to complement such efforts and assist the stakeholders (e.g., researchers, journals, peer reviewers, funders, policymakers) in assessing and improving research output efficiently. In this talk, I will discuss how NLP methods can help improve rigor and transparency in biomedical communication, as reflected in textual artifacts such as protocols, manuscripts, and publications. I will then present two research projects that we have been pursuing in this area: a) evaluating the methodological and reporting quality of clinical trial publications, and b) assessing citation integrity in biomedical literature. Finally, I will highlight some of the key challenges for NLP in this application domain.

Speaker Bio:
Dr. Halil Kilicoglu is Associate Professor in the School of Information Sciences (iSchool) at the University of Illinois Urbana-Champaign. He specializes in natural language processing (NLP), artificial intelligence/machine learning (AI/ML), and knowledge representation with biomedical applications. His work aims to extract and organize knowledge buried in textual artifacts to benefit biomedical discovery and scholarship, and improve healthcare outcomes. His recent work includes development of automated methods to assess research rigor, transparency, and integrity of biomedical publications. Prior to joining the iSchool faculty in 2019, he was a research scientist at the U.S. National Library of Medicine, National Institutes of Health (NLM/NIH), where he led the Semantic Knowledge Representation project. His research has been funded by NIH, AHRQ, and HHS Office of Research Integrity.

BIME 591
Wednesdays – 11:30-12:20 pm
Section B, HSEB 421
Zoom Information: https://washington.zoom.us/my/velvinfu?pwd=Y3dJbjNBeFpzTC9HTXV1UDFYYXlKQT09

ANNOUNCEMENTS
UPCOMING WORKSHOP
Considering Cultural and Linguistic Diversity in AI Applications: A Hybrid Workshop

How are people affected differently by AI?
How might aspects of one’s culture, background, and other characteristics or traits affect how we might respond to AI?
How might these considerations inform how we regard or develop applications involving AI?
How do we collaboratively work towards integrating the values and preferences of communities of interest in AI tools?

Event purposes:

Raise awareness about the importance of thoughtful engagement in the design of AI applications for culturally and linguistically diverse populations;
Offer a venue to engage in activities to think about relevant concepts and build relevant skills; and
Present and hear about the work of others on this topic.

Workshop locations: University of Washington (UW); University of California, Irvine (UCI); and virtual.
Event Sponsors:

Association for Information Science & Technology (ASIS&T) U.S. West Chapter
UCI Connected Learning Lab

2025 UW Wellness Symposium
Enhance your well-being at the 2025 UW Wellness Symposium, Wednesday, May 21, and Thursday, May 22. Join expert-led virtual sessions on mental, physical and financial health with The Whole U.

April 14, 2025 – April 18, 2025

UPCOMING LECTURES AND SEMINARS
BIME 590
Presenter: Daniel S. Weld, PhD
Thursday, April 24^th – 11-11:50 am
850 Republican Street, Building C, Room 123 A/B
Zoom Information: https://washington.zoom.us/my/bime590

Speaker will present in-person

Title:
Intelligence Augmentation for Scientific Researchers

Abstract:
Recent advances in Artificial Intelligence are powering revolutionary interactive tools that will transform the very nature of the scientific enterprise, leading to increasingly automated scientific discovery. We describe several large-scale projects at the Allen Institute for AI aimed at developing open models, agentic platforms, and novel interaction that amplify the productivity of scientists and engineers.

Speaker Bio:
Daniel S. Weld is Chief Scientist and General Manager of Semantic Scholar at the Allen Institute of Artificial Intelligence and Professor Emeritus at the University of Washington. After formative education at Phillips Academy, he received bachelor’s degrees in both Computer Science and Biochemistry at Yale University in 1982. He landed a Ph.D. from the MIT Artificial Intelligence Lab in 1988, received a Presidential Young Investigator’s award in 1989, an Office of Naval Research Young Investigator’s award in 1990; he is a Fellow of the Association for Artificial Intelligence (AAAI), the American Association for the Advancement of Science (AAAS), and the Association for Computing Machinery (ACM). Dan was a founding editor for the Journal of AI Research, was area editor for the Journal of the ACM and on the editorial board for the Artificial Intelligence journal. Weld is a Venture Partner at the Madrona Venture Group and has co-founded several companies, including Netbot (sold to Excite), Adrelevance (sold to Media Metrix), and Nimble Technology (sold to Actuate).

BIME 591
Wednesdays – 11:30-12:20 pm
Section B, HSEB 421
Zoom Information: https://washington.zoom.us/my/velvinfu?pwd=Y3dJbjNBeFpzTC9HTXV1UDFYYXlKQT09

PAPERS & PUBLICATIONS
Luna Xingyu Li, Ray-Yuan Chung, Feng Chen, Wenyu Zeng, Yein Jeon, Oleg Zaslavsky, “Learning from Elders: Making an LLM-powered Chatbot for Retirement Communities more Accessible through User-centered Design”, accepted by the CALD-AI workshop “Considering Cultural and Linguistic Diversity in AI Applications” at ASIS&T.

ANNOUNCEMENTS
2025 UW Wellness Symposium
Enhance your well-being at the 2025 UW Wellness Symposium, Wednesday, May 21, and Thursday, May 22. Join expert-led virtual sessions on mental, physical and financial health with The Whole U.

April 7, 2025 – April 11, 2025

UPCOMING LECTURES AND SEMINARS
BIME 590
Presenter: Dr. Wanda Pratt, PhD, FACMI
Thursday, April 17^th – 11-11:50 am
850 Republican Street, Building C, Room 123 A/B
Speaker will present in-person

Title:
Informatics to Support Inclusive and Equitable Healthcare

Abstract:
Many people face marginalization by today’s healthcare system or society at large, and those experiences often lead to healthcare inequities as well as poor health outcomes. Marginalization can arise from a variety of sources, including social media, clinical encounters, health research, and even health technologies. Recently, more attention has been paid to these issues, such as through weekly highlights of equity, diversity, and inclusion articles in the high-profile medical journal JAMA; however, much more work is needed. Informatics approaches could provide new opportunities both to understand these problems more deeply and to develop new systems and strategies to reduce the problems. In this talk, I will describe several informatics projects that utilize person-centered, participatory, or community-based approaches to understand people’s experiences and to support more inclusive and equitable healthcare.

Speaker Bio:
Dr. Wanda Pratt is a Professor in the Information School with an adjunct appointment in Biomedical Informatics & Medical Education in the Medical School at the University of Washington. She also served as the Information School’s inaugural Associate Dean for Inclusion, Diversity, Equity, Access, and Sovereignty (IDEAS). She received her Ph.D. in Medical Informatics from Stanford University, her M.S. in Computer Science from the University of Texas, and her B.S. in Electrical Engineering from the University of Kansas. Her research focuses on both understanding the work people do to manage their health as well as designing new technologies to support that work and reduce its burden. She has worked with hospitalized patients as well as people coping with a variety of chronic diseases, such as cancer, diabetes, asthma, and heart disease. Her recent work focuses on support for people from historically marginalized or underestimated communities. Dr. Pratt has received best paper awards from the American Medical Informatics Association (AMIA), the ACM CHI Conference on Human Factors in Computing Systems, the ACM Conference on Computer-Supported Cooperative Work (CSCW), and the Journal of the American Society of Information Science & Technology (JASIS&T). Her research has been funded by the National Science Foundation, the National Institutes of Health, the Agency for Healthcare Research & Quality, the Robert Wood Johnson Foundation, Intel, Google, and Microsoft. Dr. Pratt is a fellow of the American College of Medical Informatics.

BIME 591
Wednesdays – 11:30-12:20 pm
Section B, HSEB 421
Zoom Information: https://washington.zoom.us/my/velvinfu?pwd=Y3dJbjNBeFpzTC9HTXV1UDFYYXlKQT09

PAPERS & PUBLICATIONS

Suchismita Naik, Amanda Snellinger, Austin L. Toombs, Scott Saponas, and Amanda K. Hall. 2025. Exploring Early Adopters’ Use of AI Driven Multi-Agent Systems to Inform Human-Agent Interaction Design: Insights from Industry Practice. In Extended Abstracts of the CHI Conference on Human Factors in Computing Systems (CHI EA ’25), April 26-May 1, 2025, Yokohama, Japan.
Emma McDonnell
Trump’s research cuts threaten to set off wave of university brain drain // Shelly Sakiyama-Elbert, bioengineering; Emma McDonnell, student – Washington State Standard
Selected as a Finalist for the Best Full Paper Award for iConference 2025:
Chassanoff, A., & Chen, A. T. (2025). Conceptual approaches to information-as-potentiality. iConference 2025. Bloomington, Indiana, USA.
Su, Y., Chen, A. T.,Kaneshiro, J., Domoto-Reilly, K., & Zaslavsky, O. (accepted). Pilot quasi-experimental single-arm study of a virtual intervention for caregivers of persons with Lewy body dementia. Aging and Mental Health. DOI: 1080/13607863.2025.2462758
Chen, A. T.,& Tsai, T.-I. (accepted). A narrative review on immigrant information and healthcare service utilization frameworks. Poster to be presented at MEDINFO 2025, Aug. 9-13, 2025. Taipei, Taiwan.
Babitts, A., Chen, A. T.,Fillipo, R., Friedly, J. L., Jarvik, J. G., Suri, P., Rundell, S.D. (accepted). Comparing influential variables for low and high recovery expectations among those with lumbar spinal stenosis: a network analysis. Poster to be presented at S. Association of the Study of Pain Scientific Meeting (USASP), April 29-May1, 2025: Chicago, IL.
Chien, S. Y., Zaslavsky, O., Belza, B., & Chen, A. T.(accepted). Loneliness, technology, and culture: Insights from older Chinese immigrants’ experiences. To be presented at Sigma Theta Tau International Nursing Research Congress, July 17-20, 2025. Seattle, WA, USA.

ANNOUNCEMENTS

Name Pronunciation Now Available in Workday – Take a moment to add a voice recording and the phonetic spelling of your name to your Workday profile to help your colleagues say your name correctly.

In the Workday Search field, enter “Change My Name Pronunciation.”
A pop-up window may appear asking you to allow “wd5-impl.workday.com” to use your microphone, which you’ll need to do if you wish to share a voice recording.
Follow the guidance on the screen to record a voice recording and type in the phonetic spelling of your name.
Once you select Submit, your name pronunciation will appear under your name on your Workday Profile page.

To learn how to pronounce your colleague’s name:

In the Workday Search field, enter their name, then select their name from the drop-down list
If they’ve entered their name pronunciation, you can find it under their name on the upper left corner of their Workday Profile page

Learn more on the Name Pronunciation Tools Available in Workday page.

PSO Annual Forum: Resilience in the community

Join us for workshops and discussions fostering resilience, connection, well-being and meaningful change at the UW on Tuesday, April 22, and Wednesday, April 23.

Virtual wellness series

Support your well-being and connect with the UW community through The Whole U Wellness Connection Series 2025 — a monthly, virtual gathering designed to inspire reflection, growth and self-care.

March 24, 2025 – March 28, 2025

UPCOMING LECTURES AND SEMINARS
BIME 590 More Details to Come
Thursday, April 3^rd – 11-11:50 am
850 Republican Street, Building C, Room 123 A/B
Zoom Information: https://washington.zoom.us/my/bime590

BIME 591
Wednesdays – 11:30-12:20 pm
Section B, HSEB 421
Zoom Information: https://washington.zoom.us/my/velvinfu?pwd=Y3dJbjNBeFpzTC9HTXV1UDFYYXlKQT09
Selected Topics in Natural Language Processing for Healthcare

Overview:
Recent advancements in large language models (LLMs) have demonstrated great capabilities in in-context learning, multimodal and long-context document understanding, and chain-of-thought reasoning. These capabilities are set to revolutionize healthcare, impacting areas such as patient education, automated documentation, medical error correction, and diagnostic support. In this research seminar, we will explore key papers that address the modeling, applications, and evaluation of LLMs in healthcare. This session is open to students of all levels of expertise in LLMs.

Oliver Bear Don’t Walk and Danner Peter will be giving a talk at the University of Arizona
Tuesday, April 1^st – 11-12 pm MST
Overview ⋮ CB2 Seminar – Oliver Bear Don’t Walk and Danner Peter ⋮ The University of Arizona ⋮ Events

Collaboratively Identifying Population-Specific Social Drivers of Health for Indigenous Patients Living with HIV

PAPERS & PUBLICATIONS

Changye Li, Weizhe Xu, Serguei Pakhomov, Ellen Bradley, Dror Ben-Zeev, and Trevor Cohen. Bigger But Not Better: Small Neural Language Models Outperform Large Language Models in Detection of Thought Disorder. In Proceedings of the 10th Workshop on Computational Linguistics and Clinical Psychology, NAACL 2025
Changye Li, Zhecheng Sheng, Trevor Cohen, and Serguei Pakhomov. “Is There Anything Else?”: Examining Administrator Influence on Linguistic Features from the Cookie Theft Picture Description Cognitive Test. In Proceedings of the 14th Workshop on Cognitive Modeling and Computational Linguistics. NAACL 2025

Defensive Responses to Implicit Association Tests and Bias Awareness in an Implicit Bias Mitigation Training, Janice Sabin, PhD, Michele Pugnaire, MD, Joanne Calista, MSW, Nancy Esparza, MEd, Olga Valdman, MD, Maria Garcia, MD, MPH, Majid Yazdani, MD, Janet Hale, PhD, RN, FNP, Jill Terrien, PhD, ANP-BC, Ethan Eisdorfer, PsyD, Valerie Zolezzi-Wyndham, JD, Germán Chiriboga, MPH, Geraldine Puerto, MPH, Stacy Potts, MD, Sylvia Stanhope, BS, Jeroan Allison, MD, MS, Vennesa Duodu, BS, BA, Jennifer Tjia, MD, MSCE, (2025) Advances in Medical Education and Practice, 2025:16 419–430.

Award:
X. Zhang, G. Luo, and S.B. Zeliadt. Transforming Quality Measurement – Integrating Information from Narrative Clinical Notes to Augment Structured Quality Measures. Best abstract for the Measuring Safety, Quality and Value theme of AcademyHealth’s 2025 Annual Research Meeting, Minneapolis, MN, Jun., 2025.

ANNOUNCEMENTS
Name Pronunciation Now Available in Workday – Take a moment to add a voice recording and the phonetic spelling of your name to your Workday profile to help your colleagues say your name correctly.

In the Workday Search field, enter “Change My Name Pronunciation.”
A pop-up window may appear asking you to allow “wd5-impl.workday.com” to use your microphone, which you’ll need to do if you wish to share a voice recording.
Follow the guidance on the screen to record a voice recording and type in the phonetic spelling of your name.
Once you select Submit, your name pronunciation will appear under your name on your Workday Profile page.

To learn how to pronounce your colleague’s name:

In the Workday Search field, enter their name, then select their name from the drop-down list
If they’ve entered their name pronunciation, you can find it under their name on the upper left corner of their Workday Profile page

Learn more on the Name Pronunciation Tools Available in Workday page.

The next UW Medicine Town Hall will be on March 28 at 9:30 to 10:30 a.m. by Zoom, https://washington.zoom.us/j/98151483608. Please use this form to share your questions so they can address common themes: Submit Your Questions. Following the Town Hall, a link will be shared to the recording so those unable to attend can still view it. Please remember to check the UW Medicine Federal Policy Updates site for updates in real time.

March 10, 2025 – March 14, 2025

UPCOMING LECTURES AND SEMINARS
BIME 590 – On break until April 3^rd!

PAPERS & PUBLICATIONS
Serena Xie – Our paper titled “Patient and clinician acceptability of automated extraction of social drivers of health from clinical notes in primary care” got accepted to JAMIA. Authors: Serena Jinchen Xie, Carolin Spice, Patrick Wedgeworth, Raina Langevin, Kevin Lybarger, Angad Singh, Brian Wood, Jared Klein, Gary Hsieh, Herbert Duber, Andrea Hartzler

Serena Xie – Our paper titled “Adapting Communication Styles in Health Chatbot using Large Language Models to Support Family Caregivers from Multicultural Backgrounds” is accepted to CHI 2025. Authors: Rebekah Baik, Stephanie Lee, Serena Jinchen Xie, Wang Liao, Elina Hwang, Weichao Yuwen.

ANNOUNCEMENTS
Please join us in congratulating Chak Charoensilpchai who successfully passed his General Exam!

Title: Assessing the Acceptability of the Mobile Health Application “Sor Or Nor Buddy” for Community Health Workers in Thailand’s Primary Care Settings

Abstract: Mobile health (mHealth) technologies are essential in low- and middle-income countries (LMICs) due to their affordability and accessibility. In Thailand, the Ministry of Public Health introduced Sor Or Nor Buddy, a mobile application designed to support community health workers (CHWs) in primary care via smartphones and tablets. However, its development was driven through a top-down approach, with minimal input from end users, particularly CHWs, raising concerns about its alignment with user needs. Moreover, evaluations conducted by the Ministry of Public Health focus primarily on technical functionality and system workflows, while CHWs’ acceptance remains understudied.

This study aims to assess the app’s acceptability using the Unified Theory of Acceptance and Use of Technology (UTAUT). Aim 1 conducts a scoping review of mHealth interventions in LMICs to identify trends and benchmarks for understanding health workers’ acceptance, aligned with the Thai context. Aim 2 develops a validated and reliable survey instrument to quantitatively assess CHW acceptance. Aim 3 integrates quantitative and qualitative methods to assess acceptance in primary care settings. Findings will enhance understanding of mHealth adoption in LMICs, inform CHW-centered digital health strategies, and contribute to global mHealth implementation frameworks.

Please join us in congratulating Yile Chen who successfully passed her General Exam!

Title: Advancing Variant Interpretation: A Gene-Specific Framework for Prioritization, Prior Estimation, and Calibration to Enhance Evidence Strength and Clinical Significance Classification

Abstract: Genome medicine relies on the accurate classification of genomic variants to guide clinical decision-making, yet a significant proportion of variants, particularly missense variants, remain classified as variants of uncertain significance (VUS). Functional assays and computational predictors are critical for generating evidence to resolve VUS, but challenges persist in prioritizing genes for functional studies and calibrating computational predictions for clinical use. This dissertation proposes advanced algorithms to enhance variant classification through gene prioritization and predictor calibration. In Aim 1, I will develop a framework to prioritize genes for functional assays by integrating clinical utility, potential to improve computational predictors, and assay feasibility. This framework will optimize resource allocation and enhance predictive accuracy. In Aim 2, I will estimate gene-specific prior probabilities of pathogenicity using positive-unlabeled learning, leveraging data from ClinVar and population databases. This will provide the first comprehensive curation of gene-specific priors, enabling improved variant classification. In Aim 3, I will calibrate gene-specific variant effect predictors (VEPs) to align with ACMG/AMP guidelines, ensuring consistent interpretation of VEP scores in clinical settings. Together, these aims will advance the accuracy and clinical applicability of variant classification, addressing critical challenges in genome medicine and improving patient care

March 3, 2025 – March 7, 2025

UPCOMING LECTURES AND SEMINARS

BIME 590
Presenter: Dr. Patrick Wedgeworth, MD, MISM
Thursday, March 13^th – 11-11:50 am
850 Republican Street, Building C, Room 123 A/B
Zoom Information: https://washington.zoom.us/my/bime590

Speaker will present in-person

Title:
Using Geospatial Data to Determine the Role of Place in Audiovisual Hallucinations

Abstract:
Based on a dataset generated for prior work for Dr. Trevor Cohen and Dr. Dror Ben-Zeev, as part of a new U01 grant with these principal investigators, our team will leverage geospatial data from study participants across the country, ecological momentary assessments, and information from Google Maps to understand the relationship between location characteristics and audiovisual hallucination symptoms.

Speaker Bio:
Dr. Wedgeworth is a graduate of the Clinical Informatics fellowship at University of Washington, and an Internal Medicine trained physician with a master’s degree in information systems management from Carnegie Mellon University. He has experience in Epic building, data analytics, patient care in underserved communities, and working in an advisory role for municipal public health initiatives. His work is currently focused at the intersection of clinical care, analytics and social determinants of health. He is interested in the use of data analytics and clinical decision support to drive evidence-based, socially informed care that improves patient health and outcomes

HQSC Presents
Wedneday, March 19, 2025 6:30pm Pacific Time
Presentor: Dr. Shak Rheman, Interim Chair and Professor of Bioinformatics and Internal Medicine at University of Arizona
Title: Leveraging Negotiation & Communications Skills for Success

Our UW House Quality and Safety Committee (HQSC) has an evening meeting with a bioinformatics faculty from Arizona speaking about negotiation and communication skills. I wanted to pass this meeting invite along to any BIME students and staff that would like to attend. – Jason Lau

PAPERS & PUBLICATIONS
Baumgartner, A., Robinson, M., Ertekin-Taner, N. Golde TE, Jaydev S, Huang S, Hadlock J & Funk C. Fokker-Planck diffusion maps of microglial transcriptomes reveal radial differentiation into substates associated with Alzheimer’s pathology. Communications Biology. 2025.

https://doi.org/10.1038/s42003-025-07594-y

X. Zhang, G. Luo, and S.B. Zeliadt. Transforming Quality Measurement – Integrating Information from Narrative Clinical Notes to Augment Structured Quality Measures. AcademyHealth’s 2025 Annual Research Meeting, Minneapolis, MN, Jun., 2025.

X. Zhang, J. Douglas, C. Eck, G. Luo, B. Bokhour, and S.B. Zeliadt. Associations between Initiation of Whole Health Services Utilization and Quality Measures: A Population-Level Retrospective Comparative Study in the Veterans Health Administration. AcademyHealth’s 2025 Annual Research Meeting, Minneapolis, MN, Jun., 2025.

X. Zhang, J. Douglas, C. Eck, G. Luo, B. Bokhour, and S.B. Zeliadt. Associations between Initiation of Whole Health Services Utilization and Quality Measures: A Population-Level Retrospective Comparative Study in the Veterans Health Administration. SGIM (Society of General Internal Medicine) 2025 Annual Meeting, Hollywood, FL, May, 2025.

S. Zeng, S.S. Coggeshall, E.W. Rosser, S.L. Taylor, D.J. Burgess, G. Luo, and S.B. Zeliadt. Population-Level Tobacco Cessation Outcomes Associated with Implementing Whole Health at the Veterans Health Administration. AcademyHealth’s 2025 Annual Research Meeting, Minneapolis, MN, Jun., 2025.

UPCOMING EXAMS
Title: Advancing Variant Interpretation: A Gene-Specific Framework for Prioritization, Prior Estimation, and Calibration to Enhance Evidence Strength and Clinical Significance Classification
Student: Yile Chen
Date/Time: Tuesday, March 11, 2025, at 3:00 PM (Pacific Time)
Exam Location: In person(Health Science Building H-wing H562)
Zoom: https://washington.zoom.us/j/93190437065

ANNOUNCEMENTS
40 Under 40 – Now in its 27th year, the Puget Sound Business Journal recognizes 40 people each year who are under the age of 40 who are making a difference in their companies and communities. This year, we are proud to announce that Dr. Angad Singh is being recognized as one of those leaders and now everyone in the region will know what an amazing person Angad is and what he does for our community. Congratulations, Angad!

Please join us in congratulating Brian Chang who successfully passed his PhD defense on February 28^th!

Title: Leveraging multimodal models to detect osteoporotic compression fractures

Abstract: Osteoporosis is a chronic disease of low bone mineral density that affects older patients, predisposing them to fractures. While osteoporosis screening is evidence-based, it remains grossly under-utilized. Osteoporotic compression fractures (OCFs) are an early biomarker for osteoporosis but are often misclassified and under-reported on review by radiologists. Opportunistic screening, or leveraging pre-existing data, to detect OCFs to augment the current standard of osteoporosis screening could prompt appropriate diagnostic studies, treatment, and risk management. Current fracture detection tools show promise but are limited by key factors, namely manual curation of data inputs and lack of external validation and generalizability, limiting their potential clinical utility. They are also based on unimodal models, or those that leverage a single data modality. Multimodal models that leverage more than one modality have shown improved performance in clinical tasks and also better reflect real-world clinical workflows.

In this dissertation, I focus on developing and evaluating multimodal models to detect OCFs, leveraging unstructured clinical notes, radiographs, and structured electronic health record (EHR) data. To achieve this, a spine radiograph dataset from previous work in our group was used. Matching patient IDs from this dataset, I obtained clinical notes from a quaternary healthcare enterprise database to annotate fracture events. With these datasets, I implemented and evaluated unimodal models for each of the modalities above (images only and notes only) to produce outputs for the multimodal models, described in the following aims in this research:
1) Aim 1: Implement and evaluate transformer models to extract fracture events from clinical notes. An ensemble algorithm to consolidate fracture events at the note- and patient-level was also developed to produce both structured data representing a patient history of fractures and feature representations for downstream input separately to multimodal models (Aim 3). Evaluation metrics demonstrated that fine-tuned transformer models are able to extract fracture events from clinical notes with good performance, albeit limited by the small training corpus.
2) Aim 2: Develop and assess an imaging analysis pipeline for detecting OCFs. An imaging analysis pipeline consisting of independently developed machine learning models were chained in a fully automated framework. Evaluation of the pipeline was performed with a dataset of radiographs acquired in various clinical settings to measure real-world performance. While we were able to develop a performant fully automated pipeline, the evaluation demonstrated subpar performance in detecting positive cases for OCFs at the image-level.
3) Aim 3: Develop and assess whether multimodal models combining NLP, imaging analysis, and structured EHR data perform better than imaging analysis alone in detecting OCFs. With the structured data and feature representations from Aim 1, the imaging analysis pipeline predictions from Aim 2, and other structured EHR data, numerous multimodal model architectures were trained and evaluated in detecting OCFs at the patient-level. The evaluation of these models demonstrated better performance than the unimodal models (images and notes only) in detecting OCFs even with a small training corpus, reaching an acceptable absolute performance.

Please join us in Congratulating YiFan Wu who successfully passed her PhD defense on March 3^rd!
Title: Leveraging temporality, dose effect, and co-medication to improve drug safety surveillance

Abstract: Adverse drug reactions (ADRs) rank among the top causes of morbidity and mortality worldwide, yet current post-market drug surveillance systems often relying on spontaneous reporting. They suffer from under-reporting of ADRs and limited capture of clinical context. This dissertation addresses these gaps by leveraging electronic health record (EHR) data and transformer-based models to detect ADRs and drug–drug interactions (DDIs) more effectively.

First, we develop and evaluate a generative transformer architecture (GPT-2) trained from scratch on longitudinal EHR data from two distinct repositories (MIMIC-IV and UW). Unlike traditional disproportionality metrics that focus on cross-sectional drug-event co-occurrences, the proposed model captures temporal relationships and contextual dependencies among medications, diagnoses, and outcomes. Second, we introduce a “value-aware” embedding approach to incorporate continuous numeric data, such as drug dosages and lab measurements. Experimental results show that these value-aware embeddings further improve model performance, outperforming baseline transformer architectures that did not have numeric data. Third, we extend the model’s scope to evaluate DDIs under polypharmacy conditions, demonstrating that a transformer exceeded the predictive accuracy of simpler machine learning baselines.

February 24, 2025 – February 28, 2025

UPCOMING LECTURES AND SEMINARS
BIME 590
Presenter: Anthony (Tony) Rossini, ScD
Thursday, March 6^th – 11-11:50 am
850 Republican Street, Building C, Room 123 A/B
Zoom Information: https://washington.zoom.us/my/bime590

Speaker will present Remotely

Title:
Quality in Evidence, Quality Evidence, Quality of Evidence: Data Governance, Data Science, and human experiences as a basis for trust in our evidence

Abstract:

Evidence, whether as generated in a public institution or private corporation, is one basis for decision-making and therefore quality evidence is treasured. Belief in the Garbage-In-Garbage-Out principle leads to a desire for quality in the inputs. There are 3 facets to the inputs: quality for data is manifested in Data Governance, quality for processing data is manifested in Data Science, and quality for Operational Processes are the glue that provide some assurance that both are done at an acceptable level. We take a brief journey through all 3 facets, which I’ve worked on across the years, ending with directions for future work. There will be an obligatory reference to Artificial Intelligence.

Speaker Bio:
Anthony (Tony) Rossini is currently Senior Data Office Governance Lead at UCB Pharma. During the last 3 years, he has worked in the Data Office at UCB supporting a range of enterprise-wide multiscale data governance initiatives. Prior to that, he worked for 17 years in a range of data science domains at Novartis, in both expert and manager roles, including pharmacometrics, medical diagnostic R&D, statistical methodology, scientific oversight, quantitative safety, and real world evidence. Side roles in all positions focused on the interplay between data science, data, IT platforms, and processes. He was a research faculty member at UW at MEBI and Biostatistics before that with research interests in the data science user experience, including seminal work with IDEs, virtual reality, parallel computing, visualization, and literate programming.

PAPERS & PUBLICATIONS
Sriram, V., Conard, A. M., Rosenberg, I., Kim, D., Saponas, T. S., & Hall, A. K. (2025). Addressing biomedical data challenges and opportunities to inform a large-scale data lifecycle for enhanced data sharing, interoperability, analysis, and collaboration across stakeholders. Scientific Reports, 15(1), 6291. https://doi.org/10.1038/s41598-025-90453-x

Kim KK, Backonja U. Digital health equity frameworks and key concepts: a scoping review. J Am Med Inform Assoc. Published online February 12, 2025. doi:10.1093/jamia/ocaf017
It is free to download here:
https://academic.oup.com/jamia/advance-article/doi/10.1093/jamia/ocaf017/8010284

UPCOMING EXAMS
Title: Leveraging multimodal models to detect osteoporotic compression fractures
Student: Brian Chang
Date/Time: Friday, February 28, 2025 at 8:00AM Pacific Time
Zoom Only: https://washington.zoom.us/my/peter.th

Title: Leveraging temporality, dose effect, and co-medication to improve drug safety surveillance
Student: YiFan Wu
Date/Time: Monday, March 3, 2025, at 1:00 PM (Pacific Time)
Exam location: In person(SLU Building E130B)
Zoom: https://washington.zoom.us/my/cohenta

News and Events

Chair’s Message

Biomedical Informatics and Medical Education Newsletter

April 28, 2025 – May 2, 2025

April 21, 2025 – April 25, 2025

April 14, 2025 – April 18, 2025

April 7, 2025 – April 11, 2025

March 24, 2025 – March 28, 2025

March 10, 2025 – March 14, 2025

March 3, 2025 – March 7, 2025

February 24, 2025 – February 28, 2025

Upcoming Events