Biography
Jinying Chen, PhD, is an assistant professor of Department of Medicine/Section of Preventive Medicine and Epidemiology and a faculty member of Data Science Core at Boston University Chobanian & Avedisian School of Medicine. She also serves as the chair of the Action Group of Technology in Implementation Science in the Consortium for Cancer Implementation Science (CCIS).
Dr. Chen has a PhD in Computer and Information Science from the University of Pennsylvania and is specialized in natural language processing (NLP) and machine learning. She received postdoctoral training in Health Sciences, including biostatistics, health informatics, and implementation science, from the University of Massachusetts Chan Medical School. Dr. Chen’s research focuses on method innovations (NLP, machine learning/deep learning, big data analytics, and biostatistics) to support public health and health services research. Her research areas include:
NLP and Machine Learning for Health Services Research. Dr. Chen developed and user tested NoteAid, the first NLP system that automatically links medical terms in electronic health record (EHR) notes to lay-language definitions to help patients comprehend their notes. She applied a variety of machine learning techniques, including tree-based and graph-based ensemble learning and transfer learning, to support the development of NoteAid. Working with physician scientists, she developed NLP systems and methods to extract information (e.g., COVID-19 symptoms, pain symptoms, hypoglycemia events, cognitive test results) from EHR free text to support epidemiology studies and the creation of a national cohort of patients with Alzheimer’s disease. She is the 2023 recipient of the LTC Data Cooperative's Real-World Data Scholarship and is leading an initiative to validate billions of medication records in the LTC dataset using NLP and large language models (LLMs).
Risk Prediction, Digital Biomarkers, and Deep Phenotyping for Brain Health. Since joining BU, Dr. Chen has been working with other BU faculty members to develop and apply advanced data science methods to brain health research. She is the recipient of the 2023 pilot award from the Framingham Heart Study Brain Aging Program (FHS-BAP) to lead the efforts in developing a new construct-based approach to improve the interpretability and generalizability of machine learning-based risk prediction for Alzheimer’s disease (AD). Other work includes developing new methods that leverage LLMs to support biomedical data harmonization, investiging social and societal factors influencing AD progression and care, applying deep learning techniques for AD phenotyping and progression prediction, and developing speech markers for AD risk.
Innovation in Implementation Science Methods. As a K12 scholar in implementation science, Dr. Chen developed implementation strategies to support pain assessment in cardiovascular patients post discharge and used NLP to enhance patient recruitment and statistical analysis. As the recipient of a pilot award from the NCI-funded iDAPT Implementation Science Center for Cancer Control, she developed EHR-based metrics and machine learning methods (e.g., unsupervised statistical latent-variable learning models) to identify clinical activity patterns from EHR logs and successfully applied this approach to monitor the impact of tobacco cessation tools implemented in cancer clinics. Her approach integrated methods from biostatistics, machine learning, and NLP.
Digital Health Interventions and Tools. Dr. Chen collaborated with researchers from University of Massachusetts Chan Medical School, Wake Forest School of Medicine, Boston University, and Implementation Science Centers for Cancer Control (ISC3) to study the effects of digital health interventions and tools. She has led or contributed to projects that (1) assessed longitudinal behavior change in people who smoke following digital health interventions, using statistical methods such as generalized estimating equation models, time-to-event analysis, and multiple imputation to handle missing data; (2) assessed users’ adoption and engagement with digital health interventions in a variety of settings, including cognitive assessment testing, smoking cessation, transitional care, tobacco screening, and cancer screening.