micromotion¶
Analysis of human micromotion in motion time series: optical marker data, body-worn accelerometers, and force-plate centre of pressure.
The measure the package exists for is quantity of motion — the average speed of a body part, band-limited to 0.2–10 Hz, in millimetres per second. It applies equally to all three sensor families because the shared abstraction is the frequency band, not the instrument.
Why it exists¶
Across the research repository this was extracted from there were 159 analysis scripts. Fifty-eight defined their own band-pass filter and thirty-seven computed quantity of motion. The project's central measure existed in dozens of copies that did not all agree, and the disagreements were invisible: two scripts returned 3.77 and 4.20 mm/s for the same series purely because their filter chains differed, and neither said so.
That is the argument for a library. Not tidiness — reproducibility of the central quantity.
What it will not let you do¶
Most of the design is about refusing comparisons that look reasonable and are not.
- Asking for the legacy optical band on accelerometer data raises, because gravity is a DC term and an accelerometer cannot be integrated without a lower edge.
- Upsampling raises, because it invents structure that scale-sensitive methods then treat as real.
- Gap sentinels become
NaNon read, because a marker "at the origin" and a centre of pressure at the exact middle of a board are plausible-looking values that are not measurements. - Partial and filter-contaminated bins are flagged, not silently included.
Where to start¶
| If you want to | Read |
|---|---|
| compute a quantity of motion | Getting started |
| know which filter convention to use | The two bands |
| compare across datasets | Sampling rates |
| load a file | Reading files |
| combine with MGT or ambiscape | Working with other packages |
| look up a function | API reference |