Collaboration Stories
The purpose of this series is to tell the story of long-lasting, as well as emergent, stories of collaboration among clinicians and methodologists, content experts, and engineers, who have successfully worked on joint projects at the intersection of AI and health. Such collaborations are at the center of e-HAIL’s mission, and we offer these stories as a way of inspiring collaboration in other researchers.
e-HAIL Supports Development of New Essential Diagnostics Management Tool
Ye Chan Kim, MSI, M.Sc.
e-HAIL Programmer
Lee Schroeder, M.D., Ph.D.
Associate Professor, Chemical Pathology
Associate Director, Division of Clinical Pathology
Ye Chan Kim, MSI, M.SC., e-HAIL’s programmer, recently completed working on a new interactive diagnostic test management application with Associate Professor of Pathology Dr. Lee Schroeder. The application is based on Dr. Lee’s previous model for optimizing the distribution of essential diagnostics across different levels of healthcare facilities—primary (such as small clinics), secondary (such as small size hospitals), and tertiary (such as large hospitals) tiers. With the application, medical organizations and governmental agencies can figure out the optimal distribution of diagnostic tests and systems across a country. This model is crucial for achieving Universal Health Coverage (UHC), especially in low- and middle-income countries, where healthcare resources are often constrained.
Dr. Schroeder’s research highlights the importance of diagnostics in patient care, but also finds that they are frequently overlooked in global health strategies. Diagnostics play a critical role in disease detection, management, and treatment, yet their availability is often inconsistent across healthcare facilities. His model provides tier-specific recommendations for where certain diagnostic tests should be made available, ensuring that patients receive appropriate care at the lowest tier possible, reducing costs and improving access.
Working with Kim, Dr. Schroeder sought to take this theoretical model and turn it into a practical, user-friendly application for health policymakers and practitioners.The goal of the app is to enable users to input data from the healthcare facility that they’re responsible for, generate diagnostic test recommendations, and visualize how diagnostics are distributed across different tiers. The app facilitates easier comparisons with existing models, helping organizations ensure that diagnostic resources are optimally allocated.
Kim found translating a complex theoretical model into something tangible and user-friendly one of the most interesting aspects of this project. “It was particularly challenging to ensure that the app’s functionality remained robust while keeping the interface simple enough for non-technical users to navigate,” according to Kim. “The project really pushed me to think about how to balance complexity with accessibility, and that’s where I think we made the biggest impact.”
The application is available online and is open access. Its primary features include: (a) Allowing users to create custom tiers for diagnostics and conditions; (b) Visualizing diagnostic test tiers in an intuitive interface; (c) Providing a comprehensive summary of diagnostic tests, categorized by health system tiers; and (d) Enhancing decision-making by highlighting gaps or redundancies in diagnostic test allocation, especially in resource-constrained settings.
e-HAIL convener Akbar Waljee, M.D., M.Sc., AGAF, stresses the broader applicability of Kim’s work on Dr Lee’s project. According to him “the development of a similar app for health policy makers and/or practitioners is something that can be included in grant applications on a number of topics. It is a great third Aim for making a project’s findings accessible and usable.”
The Role of e-HAIL Programmer, Ye Chan Kim
Kim joined e-HAIL in April of this year as a programmer to support e-HAIL’s mission of positioning faculty dyads of AI and Health experts for external funding opportunities and creating shared resources in the form of data set and code. By working closely with faculty researchers, he focuses on supporting the testing of AI models and turning them into practical, effective, and reliable tools. By supporting their research, Kim not only makes new AI-based tools but also strengthens their chances to apply for grants to further develop their research.
Schroeder, for one, is very pleased with Kim’s support on the app, which will not only help health policy makers on the ground, but also position his team to secure additional external funding for this work. He notes, “Before working with the e-HAIL programmer, we had published multiple studies using the database, but had not been able to create a user-friendly interface with it, thus limiting the usefulness for its intended audiences. Ye Chan made more progress with this in a shorter time period than all previous attempts. We now have a useful decision support for Essential Diagnostics that will be available to health policy makers in Low-and Middle Income Countries, and I attribute all of this to the e-HAIL programmer. He has been extremely thoughtful, easy to work with, punctual, and efficient. It has been a pleasure!”
Ye Chan received his education from the University of Michigan with master’s degrees in Science Information and Science. Afterward, he worked at C3 AI in the Bay Area for two years, focusing on implementing and deploying deep learning models for failure prediction, recommendation engines, and machine learning (ML) interpretability integration. His skill sets include developing AI/ML pipelines and different peripheral infrastructure to continuously monitor the system’s performance as well as apply different operations to the model to maintain its optimal performance. He is also expanding his skillset to include backend and frontend development.
Kim looks forward to continuing to grow his skills and experiences. The complexity of applying AI to the health field is a subject he finds particularly interesting, and eHAIL has positioned him perfectly to explore that. He finds it fascinating to create scalable applications, refine existing tools, and make AI solutions more intuitive and user-friendly. He also looks forward to exploring new opportunities, like using EHR data for predictive models that could assist in real-time clinical decision-making.
According to e-HAIL convener Rada Mihalcea, Ph.D., having a dedicated programmer on the e-HAIL team is crucial to the program’s success. “The availability of a skilled programmer can truly empower and expand these cross-disciplinary collaborations by enabling rapid prototyping and deployment. e-HAIL teams can be more ambitious in their thinking and vision.”