About & Projects

About WIT

The Wearable and Implantable Technologies (WIT) group at Johns Hopkins develops open methods and software for using data from wearable and implantable technologies in health research. We study how sensor streams can describe everyday movement, sleep, physiology, and health, and build reproducible tools that make those analyses easier to perform and evaluate.

Our work connects data collected in daily life with rigorous statistical and clinical research. Alongside new analytical methods, we share software, documentation, and teaching materials so that research teams can inspect, reuse, and extend the work.

Featured open-source projects

These public repositories reflect the group’s work in wearable-sensor data processing, activity and sleep measurement, and related health-data research.

Wearable data and activity

Disclaimer: Many of these are adapted software for other work. Please explore the underlying work and licenses in typical use. The following projects are for research purposes only and should not be used for clinical decision-making.

• actinet — estimate human activity from accelerometry data.
• actimetrics — create metrics for actigraphy and activity analysis. Documentation
• actibase — baseline tools for actigraphy and activity analysis. Documentation
• actiread — tools for reading actigraphy and activity data. Documentation
• actisensorlog — analyze SensorLog and SensorLogger activity data. Documentation
• actipy — process wearable sensor data with actipy.
• agcounter — process ActiGraph counts.
• walking — identify walking and other activities from accelerometry data.
• stepcount — R interface to the OxWearables step-counting tools.
• cadence — compute cadence metrics from NHANES step-count data.

Sleep, physiology, and health research

• asleep — estimate sleep from accelerometry data. Documentation
• sleeper — compute sleep metrics from wrist-worn accelerometers.
• weartime — estimate device wear time from accelerometer data.
• wfdb — R interface for reading PhysioNet waveform database files.
• Hemodynamics-and-Urinary-Biomarkers — analysis code for the HARBOR prospective cohort study.
• mapnhanespa and actiquantiles — map physical-activity and normalized activity quantiles. actiquantiles documentation

Software, teaching, and data resources

• WearableR — short course for wearable data analysis in R. Course site
• accel_in_r — course materials for working with accelerometers in R. Course site
• activerse — a collection of related activity-analysis packages. Documentation
• actiwalkability — access the EPA Walkability Index. Documentation
• waterways — tools for analyzing OCEANS/WAVES data. Documentation
• svyfosr — survey-weighted and design-based functions on scalar regression. Documentation
• check_activity_dates — public project with a live site.
• nhanes_derivatives — presentation of NHANES accelerometry derivatives. Presentation site

For the full, current list of public repositories, visit github.com/jhuwit.