Skip to contents

Rcpp MAGE Calculation

mage_rcpp.Rd

Calculates Mean Amplitude of Glycemic Excursions (MAGE) with an Rcpp backend.

Usage

mage_rcpp(
  data,
  version = c("ma", "naive"),
  sd_multiplier = 1,
  short_ma = 5,
  long_ma = 32,
  return_type = c("num", "df"),
  direction = c("avg", "service", "max", "plus", "minus"),
  tz = "",
  inter_gap = 45,
  max_gap = 180
)

Arguments

data

A dataframe containing CGM data with columns id, time, and gl.

version

Either "ma" or "naive". Defaults to "ma".

sd_multiplier

Multiplier for the standard deviation threshold in version = "naive".

short_ma

Short moving-average window length. Defaults to 5.

long_ma

Long moving-average window length. Defaults to 32.

return_type

Either "num" for one metric per id or "df" for a list-column of segment-level MAGE rows.

direction

One of "avg", "service", "max", "plus", or "minus".

tz

Time zone used for day-grid alignment when supplied.

inter_gap

Maximum gap, in minutes, over which linear interpolation is allowed. Defaults to 45.

max_gap

Gap length, in minutes, above which MAGE is calculated on separate trace segments. Defaults to 180.

Value

A tibble with columns id and MAGE. With return_type = "df", MAGE is a list-column of tibbles with start, end, mage, plus_or_minus, and first_excursion.

Details

The implementation follows the MAGE calculation approaches used by iglu::mage. The default version = "ma" follows iglu's moving-average approach: CGM is interpolated to 5-minute intervals, short and long moving-average crossings identify candidate peak/nadir intervals, and countable excursions are those whose peak-to-nadir or nadir-to-peak amplitude is at least one glucose standard deviation.

The version = "naive" option matches iglu's older standard-deviation based calculation. This function is calculation-only and does not implement iglu's plotting options.

References

Service, F. John, et al. (1970). Mean amplitude of glycemic excursions, a measure of diabetic instability. Diabetes, 19(9), 644-655. doi:10.2337/diab.19.9.644

Fernandes, Nathaniel J., et al. (2022). Open-source algorithm to calculate mean amplitude of glycemic excursions using short and long moving averages. Journal of Diabetes Science and Technology, 16(2), 576-577. doi:10.1177/19322968211061165

See also

iglu::mage, conga_rcpp

Examples

library(iglu)
data(example_data_5_subject)
mage_rcpp(example_data_5_subject)
#> # A tibble: 5 × 2
#>   id         MAGE
#>   <chr>     <dbl>
#> 1 Subject 1  72.4
#> 2 Subject 2 118. 
#> 3 Subject 3 116. 
#> 4 Subject 4  70.9
#> 5 Subject 5 142. 
mage_rcpp(example_data_5_subject, version = "naive")
#> # A tibble: 5 × 2
#>   id         MAGE
#>   <chr>     <dbl>
#> 1 Subject 1  53.4
#> 2 Subject 2  78.2
#> 3 Subject 3  76.6
#> 4 Subject 4  42.9
#> 5 Subject 5  90.0