iCDCU Lab at the U of Cincinnati

Research Area #1

Clinical Informatics

This research area has two focuses with an overarching goal of streamlining clinical workflow and equipping clinicians with well-designed clinical decision support (CDS) tools to improve clinician experience and care delivery and outcomes. In the first focus (1a), clinical workflow analysis (CWA) is conducted to understand processes and bottlenecks. The workflow data come from semi-structured interviews, time and motion studies, electronic health record timestamps, and real-time locating systems. The workflow patterns will correlate with clinicians' work efficiency, job satisfaction, and burnout. The findings can inform quality improvement initiatives. In the second focus (1b), CDS tools are developed following three phases. First, machine learning models are trained and validated to predict patient outcomes retrospectively. Second, the workflow and usability of the CDS tool are explored through design research methods. Third, the CDS is implemented and its effectiveness and impact on health outcomes are evaluated in a quasi-experimental study. Both focuses involve data-driven interactive dashboards and visual analytics to facilitate the information consumption and knowledge discovery.

Selected Publications (Annotations: *Corresponding author, Wu mentee, ⁱEqual Contribution)

Wu DTY*, Barrick L, Ozkaynak M, Blondon K, Zheng K. Special Section on Workflow Automation: Principles for Designing and Developing a Workflow Monitoring Tool to Enable and Enhance Clinical Workflow Automation. Appl Clin Inform. In Press.
Liu L*, Wu DTY, Spooner SA, Ni Y. Development and Evaluation of an Automated Approach to Detect Weight Abnormalities in Pediatric Weight Charts. AMIA Annu Symp Proc. 2021 Nov. In Press.
Wu DTY, Chen AT, Manning JD, Levy-Fix G, Backonja U, Borland D, Caban JJ, Dowding DW, Hochheiser H, Kaganj V, Kandaswamy S, Kumar M, Nunez A, Pan E, Gotz D*. Evaluating Visual Analytics for Health Informatics Applications: A Systematic Review from the AMIA VIS Working Group Task Force on Evaluation. J Am Med Inform Assoc. 2019 Apr. PMID: 30840080.

Research Area #2

Providing Readable and Actionable Health Information

This research area aims to improve the accessibility, readability, understandability, and actionability of health information for both patients (or lay persons) and clinicians (or medical experts). For lay persons (2a), the strategy is to use readability assessment and/or natural language processing to simplify and summarize professional language and to facilitate the interpretation of health information. This is particularly important for people with low health literacy since their limited understanding may negatively impact their health outcomes. Conversely, the strategy for medical experts (2b) is to improve the reliability and usability of patient-generated health data (PGHD) in clinic visits, especially in the areas of mental health. By providing readable and actionable health information, both patients and clinicians will have a common ground for communication, which can improve patient-provider relationships and promote shared decision making.

Selected Publications (Annotations: *Corresponding author, Wu mentee, ⁱEqual Contribution)

Wu DTY, Hanauer DA, Mei Q, Clark PM, An LC, Proulx J, Zeng QT, Vydiswaran VV, Collins-Thompson K, Zheng K*. Assessing the Readability of Clinicaltrials.gov. J Am Med Inform Assoc. 2015 Aug. PMID: 26269536. (full-text)
Wu DTY*, Xin C, Bindhu S, Xu C, Sachdeva J, Brown JL, Jung H. Clinician Perspectives and Design Implications in Using Patient-Generated Health Data to Improve Mental Health Practices: Mixed Methods Study. JMIR Form Res. 2020 Feb. PMID: 32763884. (full-text)

Research Area #3

Supporting Clinical Research using Informatics

This research area involves developing a set of informatics tools to support data and project management as well as cohort and knowledge discovery in clinical research. The Wu lab developed an extensible and accountable data platform to meet clinicians' dynamic needs in quality improvement and clinical research; the lab has been participating in the Text Retrieval Conference (TREC) competitions since 2017 and developing information retrieval systems on datasets from PubMed and ClinicalTrials.gov. In addition, the lab contributes to the EMERSE project to help clinicians identify patient cohorts through clinical notes. Lastly, the lab works with several research groups to develop a research data system with common data models to support data analysis and hypothesis testing.

Selected Publications (Annotations: Corresponding author, Wu mentee, ⁱEqual Contribution*)

Wu DTY*, Zheng K, Bradley DJ. CHCi – A Dynamic Data Platform for Clinical Data Capture and Use. AMIA Jt Summits Transl Sci Proc. 2018 May. PMID: 29888081.
Hanauer DA*, Wu DTY, Yang L, Mei Q, Steffy KM, Vydiswaran VG, Zheng K. Development and Empirical User-Centered Evaluation of Semantically-based Query Recommendation for an Electronic Health Record Search Engine. J Biomed Inform. 2017 Jan. PMID: 28131722.

Research Area #4

Facilitating Learner-Centered Medical Education

This research is focused on the application of informatics to facilitate learner-centered approaches in medical education. In one project, the Wu lab worked with a group of medical education experts to develop a machine learning model to predict the primary resident of a patient in a care process. This prediction can connect resident performance with patient outcomes and therefore enable an outcome-based evaluation of resident training. In another project, the lab conducted a user-centered design and evaluation on an internal medicine resident dashboard to improve its utility in the competency reviews and assist learners with their improvement plan.