Outsourcing the Digital Biomarker Discovery Pipeline (2023-2024)

Background

Digital biomarkers transform data from wearables into indicators of health outcomes, allowing rapid detection, prevention and management of many diseases. To facilitate the advancement of digital health, the Digital Biomarker Discovery Pipeline (DBDP) has been created as an open-source initiative from the BIG IDEAs Lab at Duke. The objective of the DBDP is to disseminate and promote toolkits, reference methods and data as the community standard for digital biomarker development. 

However, clinicians and community members without a proficient computational background find it difficult to fully participate in the discovery of digital biomarkers. The DBDP’s current interface is not user-friendly, and the strategies for outsourcing the resources available at DBDP were limited. Therefore, DBDP must make progress in becoming an open-source community for all stakeholders. 

Project Description

Using DBDP as a case study, this project will explore and address the barriers to participating in digital biomarker discovery by exploring ways to create an evolving open-source ecosystem that can support and promote collaborative, translational research. 

Team members will create hands-on, interactive educational resources such as case studies and a graphical user interface (GUI) to serve as a visual programming tool for novices and allow users to envision how different modules can be tailored to meet specific research goals or applications.

With the prototype of the GUI, the team will survey clinicians and students to determine how to improve the GUI to increase interest in digital biomarker discovery and identify features the end-users wish to have. The team will also design case studies to address real-world problems and challenges.

Team members will strategize ways to attract researchers, clinicians and the community for involvement. They will conduct a web-based survey of the stakeholders in digital health (e.g., clinicians, researchers) to answer what motivates the developers and researchers to contribute their artifacts (data or code) to an open-source ecosystem. Based on the survey, team members will redesign the DBDP website and the GitHub repository. 

The team will also explore marketing strategies including social media and workshops to reach the community who have interests in digital medicine.

Anticipated Outputs

Interactive coding notebooks; GUI; published survey results

Student Opportunities

Ideally, this project team will include 5 graduate students and 10 undergraduate students with computational backgrounds (computer science, electrical engineering, biomedical engineering or equivalent) and/or skill sets or interests in marketing, business and communications. Students with experience in user interface and user experience design, behavioral science (survey design) and/or management are especially encouraged to apply.

All students will have the opportunity to contribute to an open-source initiative, make an impact in digital medicine and grow their connection with faculty, students and industry leaders in digital medicine. Students with a technical background will have the opportunity to design real-world application problems and propose solutions using their computational skills. They will also learn to consider practicality and the end-user in their approach. Students with a background or interest in marketing, business or communications will grow their portfolio by working with an open-source initiative in digital medicine.

Hayoung Jeong will serve as project manager. 

This project includes an optional summer component from early June through early August (2023). Students will be expected to work 12-15 hours per week on project development, including developing the GUI, designing case studies and beginning literature review.

Timing

Summer 2023 – Spring 2024

  • Summer 2023 (optional): Develop GUI; design case study problems; start market analysis and literature review; design metrics to quantify and measure success of outreach methods
  • Fall 2023: Complete case study and publish with Google Colaboratory; make case study available on repository; conduct community survey for GUI; strategize for marketing and outreach
  • Spring 2024: Refine GUI; design workshops with materials from DBDP

Crediting

Academic credit available for fall and spring semesters; summer funding available

 

Image: Logos of the BIG IDEAS Lab and the Digital Biomarker Discovery Pipeline, dbdp.org

Logos of the BIG IDEAS Lab and the Digital Biomarker Discovery Pipeline

Team Leaders

  • Bill Chen, Pratt School of Engineering–Masters of Biomedical Engineering Student
  • Jessilyn Dunn, Pratt School of Engineering-Biomedical Engineering
  • Leeor Hershkovich, Pratt School of Engineering–Biomedical Engineering–Ph.D. Student
  • Hayoung Jeong, Pratt School of Engineering–Biomedical Engineering–Ph.D. Student
  • MD Mobashir Hasan Shandhi, Pratt School of Engineering-Biomedical Engineering

/graduate Team Members

  • Mercy Guriyire, Masters of Public Policy
  • Harrison Kane,
  • Fangrui Liu, Biomedical Engineering-MS
  • Yuyou Wu, Interdisciplinary Data Science - Masters
  • Jerry Yang, Biomedical Engineering-MS

/undergraduate Team Members

  • Shun Sakai, Electrical & Computer Egr(BSE)
  • Ellie Vogel