Duke BME Master's Concentration in Health Data Science

Master's Concentration in Health Data Science


The Duke BME Master's Concentration in Health Data Science provides graduate training in the skills needed to analyze real-world health data, with a focus on interdisciplinary and team-based research and discovery.

"Duke has rich data assets, including large registries of medical images and clinical care records. We embed and mentor Duke students and clinical trainees on our teams, where they build data science and machine learning technologies that improve clinical care at Duke Health."

Suresh Balu
Program director, Duke Institute for Health Innovation (DIHI)

Through our Duke BME faculty, and our research partners associated with Duke University Hospital, the Master's Concentration in Health Data Science provides our engineering master's students with access to one of the world's leading academic medical centers.

In our courses, students gain experience with technical tools from a variety of quantitative and engineering disciplines through data-driven inquiry into contemporary health care issues.

Embeded in our partner's working teams, our students perform supervised research on large and diverse sets of clinical data while receiving meaningful mentorship from experts in the field.


Join Us

Who Should Consider This Concentration?

Duke BME engineering master's degree students who:

  • Wish to acquire state-of-the-science quantitative and programming skills
  • Desire to gain practical experience working with large clinical datasets
  • Have a personal career goal or research interest related to data science and health

Who Can Take This Concentration?

This concentration is available to students enrolled in our Master of Science (MS) and Master of Engineering (MEng) degree programs.

Learn How to Apply

To take this concentration, apply to become a Master of Science (MS) or Master of Engineering (MEng) student in Duke BME:

How to Apply


Meet Our Faculty

Primary Faculty

Sina Farsiu

Research interests: Medical imaging and image processing to improve the outcomes of patients with diseases such as age-related macular degeneration, diabetic retinopathy, Alzheimer's disease and amyotrophic lateral sclerosis (ALS).

Craig Henriquez

Research interests: Large-scale computing, heart modeling, and brain modeling.


Amanda RandlesAmanda Randles

Research interests: Biomedical simulation and high-performance computing focused on development of new computational tools to provide insight into the localization and development of human diseases.


Xiling ShenXiling Shen

Research interests: A range from systems biology and implantable devices to the study of non-coding RNA, colon cancer, stem cells, and the enteric nervous system.


Hau-Tieng WuHau-Tieng Wu

Research interests: Physiological signal processing, high-dimensional big data analysis, mathematical foundation of data analysis tools, clinical applications.

Lingchong You

Research interests: Computational systems biology and synthetic biology, including mathematical modeling of cellular networks, and mechanisms of information processing by gene networks.



Kai-Yuan ChenKai-Yuan Chen

Research interests: Systems/quantitative biology and bioinformatics, including multi-scale computational modeling and integrative analysis on omics data to reconstruct molecular signatures governing cell fate decisions, cell-cell communication, tissue homeostasis, regeneration and tumorgenesis.

Michael GaoMichael Gao

Background: A data scientist focused on applied machine learning in health care, including natural language processing and imaging — with expertise in statistical software, implementing machine learning into clinical practice, and bridging the gap between research and day-to-day operations.

Ouwen HuangOuwen Huang

Background: Co-founder and former CTO of a YCombinator image analytics company, and co-founder of a deep-learning health care company backed by the National Science Foundation. He brings expertise in entrepreneurship and leveraging tools such as Docker, Postgres, and TensorFlow.

Suyash KumarSuyash Kumar

Background: A software engineer from Silicon Valley, where he built scalable software systems leveraging technologies such as Go, Docker, Kubernetes, and Tensorflow. He co-founded a deep-learning healthcare company and is currently an associate instructor in biomedical engineering. He has a background in imaging, data science, and medical device design.

Mark Sendak

Mark Sendak

Background: The Population Health & Data Science Lead at the Duke Institute for Health Innovation — with expertise leading interdisciplinary teams to develop and deploy technologies to improve population health.


Meet Our Partners

Duke Institute for Health Innovation (DIHI)

The Duke Institute for Health Innovation (DIHI) promotes innovation in health and health care through high-impact innovation pilots, leadership development, and cultivation of a community of entrepreneurship. Learn more about DIHI

Duke MEDx: Medicine+Engineering

MEDx is an initiative that builds on the decades of collaboration between Duke's engineering and medical schools, to take our shared research and educational initiatives in exciting new directions. Learn more about MEDx

Duke Center for Applied Genomics & Precision Medicine

The Duke Center for Applied Genomics & Precision Medicine is an intellectual home for genome-inspired biomarker discovery and development, analytic and translational approaches to diagnostics, and insights into disease biology. Learn more about Duke Precision Medicine



  • BME464L – Medical Instrument Design
  • BME516 – Computational Methods in Biomedical Engineering
  • BME544 – Digital Image Processing
  • BME561L – Genome Science & Technology Lab
  • BME590L – Computational Foundations for Biomedical Simulations
  • BME790L-02 – Signal Processing and Applied Mathematics
  • BME846 – Biomedical Imaging
  • CS260 – Computational Genomics
  • CS445 – Introduction to High Dimensional Data Analytics
  • BIOSTAT 701 – Introduction to statistical theory and methods I – Intro to probability, statistical methods
  • BME 590 – Project Design
Sample BME 590 projects:
  • Use the MIMIC-III database of more than 40,000 de-identified hospital admissions to ask and answer data-driven questions
  • Compute heart rate from real ECG data; report on tachycardia and bradycardia (autocorrelation, filtering, etc) in clinically relevant cases (sepsis)
  • Develop cloud-based service to detect and “diagnose” a cardiological problem in ECG data (stenosis, heart blockages)
  • Use hidden Markov models to discover genes in the genome

Browse other Duke BME Graduate courses


Explore Duke BME Master's Degrees