Chao Chen, assistant professor in the Department of Biomedical Informatics, has won a prestigious Faculty Early Career Development (CAREER) award from the National Science Foundation (NSF). These awards support early-career faculty who have the potential to serve as academic role models in research and education, and to lead advances in the mission of their department or organization.
An expert in applications of the theory of topological data analysis in machine learning and in biomedical image analysis, Chen is also affiliated with the Department of Computer Science and the Department of Applied Mathematics & Statistics. He has been carrying out innovative, groundbreaking research in the development of topology-driven Artificial Intelligence (AI) paradigms capable of targeting many areas of science and engineering, with a particular focus on problems arising in biomedical informatics.
Chen’s NSF CAREER project, “Topology-Driven Learning for Biomedical Imaging Informatics,” will systematically incorporate topological reasoning into AI methods by leveraging the theory of persistent homology, which provides robust and differentiable representations of topology. His approach will enable topological patterns to be learned and exploited in a data-driven and task-driven manner.
Chen’s work will facilitate the use of topological information to carry out a variety of AI tasks, including to classify biomedical structures, to generate realistic synthetic data for training robust AI algorithms, and to develop methods for the representation and analysis of time-dependent data. This work is already having a tremendous impact on the field of biomedical informatics and is shedding insight into the theoretical foundation of AI. Through the course of the project, Chen will closely collaborate with domain experts to apply his methods to the extraction and analysis of complex biomedical structures arising from histopathology images and radiology images.
Chen’s project will also provide a focused data-science education to students. The education program will engage students, ranging from high school to graduate levels, through training on topology-centric data exploration with real-world data from biomedicine, science and engineering. With this hands-on approach, students will learn not only the foundations of data science, but also its practical applications.