Analytics using Axivity Sensors
At McRoberts we specialize in analyzing human mobility data from wearable sensors. In the early 1990’s we published the first papers on algorithms to classify body postures and movements. We improved and expanded our solutions ever since. Until recently, our mobility analyses could only be applied to data collected using our own DynaPort sensors. We decided to enable the use of Axivity sensors within our software ecosystem following requests from the research community. This means that you can now start measurements, upload data and analyse data using Axivity sensors on our MyMcRoberts platform in the same way our DynaPort sensors are used. With identical analysis results.
Use in IMI Mobilise-D and IDEA-FAST
DynaPort and Axivity sensors are already successfully used side-by-side in the IMI projects Mobilise-D and IDEA-FAST. MyMcRoberts is used in those large multi-center studies for data collection and processing with both sensor types.
What Prof. dr. Clemens Becker has to say on using this solution:
"At the RbMF and at the Unit of Digital Geriatrics at the University of Heidelberg we are using Axivity sensors for several observational and intervention studies. Together with these sensors we use McRoberts's data platform and analytics pipeline for processing the collected Axivity data. McRoberts's MoveMonitor algorithm is applied to the data, which provides us with a broad range of validated Digital Mobility Outcomes. Additionally, McRoberts stores all data on their servers and sends it regularly via API to our study database. The Axivity sensor fits seamlessly in McRoberts ecosystem, which is very helpful when running multi-center studies. I have been working with McRoberts with more than 10 years and am convinced that this systemic approach is what is needed for the upcoming years. I have no conflict of interest to declare. This is personal opinion."
Solutions we offer
Daily-life Mobility Analysis
Our MoveMonitor analytics package is our proven solution for analyzing daily-life mobility data. The core of the algorithm traces back to the early years of McRoberts. We have always believed in classifying physical activities with high resolution (from second to second) based on high frequency accelerometer data. This still forms the basis of our MoveMonitor analyses. Energy Expenditure estimations and Sleep Movement analyses are also part of our MoveMonitor software.
In-Depth Daily-life Gait Analysis
VU University Amsterdam developed an algorithm for calculating daily-life gait parameters. Gait speed is one of them. This method has proven itself in Fall risk prediction and is currently being tested in Knee Osteoarthritis as a marker of recovery after knee replacement. This algorithm is fully embedded in our analytics pipeline.
Axivity data can be collected using our platform and analysed instantly. We also offer the possibility to analyse existing datasets.
Making Data Available
All outcomes are stored in a database and can be shared with you in any time format, e.g. per day or week or even per minute. The file format can be chosen as well, so that you can conveniently import it in a statistical program.