Michigan Student Symposium for Interdisciplinary Statistical Sciences
04 Apr 2018 by Zhenke Wu

Tim Necamp won the Best Speed Oral Presentation at MSSISS 2018 for his work with Intern Health Study. His poster is entitled Predicting mood using multivariate mobile sensor data streams for medical interns. I also had great fun delivering junior faculty keynote talk titled Bayesian Hierarchical Methods to Power Disease Discovery and Improve Clinical Decisions. Thanks for the fantastic student organizing committee and faculty advisory committee from Biostat, EECS, IOE, Stat and Survey Methodology for showcasing the diverse statistical/data science work at Michigan.

Electronic Health Records Projects
08 Aug 2017 by Zhenke Wu

For six weeks, 44 undergraduate students from across the country met at Ann Arbor to wrestle with big data. The hands-on projects, in addition to lectures about statistics, informatics and professional development, are part of their training at 2017 Big Data Summer Institute hosted by our department. Among them, eleven students learned and applied methods to analyzing the Electronic Health Records (EHR) data. In four groups, they created their own data sets from Michigan Genomics Initiative that contains genomic information, longitudinal diagnoses, procedures and lab measurements and more to investigate heart failure, phenomewide association for lab values, Type 2 Diabetes and infectious diseases. It was great pleasure working with you!

Our team receives Data Science Challenge Initiative award in mobile health analytics
09 Apr 2017 by Zhenke Wu

Frequent interactions of individuals with mobile devices have opened new doors to behavioral and mental health research. The real-time individual-level data streams unleashed by mobile technologies have greatly improved our potential to understanding behaviors and improving health. For example, mobile technologies can capture the opportunistic windows during the day for maintaining healthy behaviors, push actionable suggestion messages, and help individuals develop and maintain long-term changes beneficial to their health. For the current project, we hope the privacy-protected data will ultimately refine our understanding of how life stress leads to depression and hence transform our ability to prevent and treat depression.

See the project award notice for more information about our project titled Identifying Real-Time Data Predictors of Stress and Depression Using Mobile Technology.

Statisticians Impacting Global Public Health
24 Feb 2017 by Zhenke Wu

There are many statistical innovations going on in Global Health. Here is the newsletter pointing to some ongoing research activities and consulting services available at Global Pubic Health, Michigan.

Teaching a special topics course about graphical models
21 Sep 2016 by Zhenke Wu

I am now teaching a PhD-level special topics course: BIOSTAT830 Statistical and Computational Methods for Learning Through Graphical Models, which will cover representation, inference, learning and causality demonstrated by case studies on real problems. Feel free to get what you need or comment on the course webpage.

Will join U of Michigan Biostat and MIDAS
05 Jul 2016 by Zhenke Wu

I will join Department of Biostatistics at Univeristy of Michigan as Assistant Professor, with joint appointment at Michigan Institute for Data Science (MIDAS), starting from September 1st, 2016.

Visualization for Individualized Health
11 May 2016 by Zhenke Wu

As part of their final project for Data Visualization for Individualized Health, Audrey Garman, Bengucan Gunen, Ruthe Huang, and Marcus Spearman, majored in public health studies, have built an excellent childhood pneumonia prediction visualization tool using Shiny. See the Hopkins inHealth and the HUB stories.

Welcome!
06 Oct 2015 by Zhenke Wu

Theory and experiment, the pillars of 19-20th century science, have been joined by computation and big data in the 21st.

I am always looking for people who take a quantitative approach to studying public health, biology and biomedicine. Particular emphasis is placed on scalable Bayesian (graphical) modeling, latent variables, causal inference and especially their intersection. If you think you may be interested, please get in touch.

I’d be happy to answer questions on research and software!