glucopy.metrics.auc

glucopy.metrics.auc(df: DataFrame, time_unit='m', threshold: int | float = 0, above: bool = True)

Calculates the Area Under the Curve (AUC) using the trapezoidal rule.

\[AUC = \frac{1}{2} \sum_{i=1}^{N-1} (X_i + X_{i+1}) * (t_{i+1} - t_i)\]

• \(X_i\) is the \(i\)-th measurement of the glucose concentration at time \(t_i\).

• \(X_{i+1}\) is the \((i+1)\)-th measurement of the glucose concentration at time \(t_{i+1}\).

• \(t_i\) is the \(i\)-th time associated with the \(X_i\) measurement.

• \(t_{i+1}\) is the \((i+1)\)-th time associated with the \(X_{i+1}\) measurement.

• \(N\) is the number of glucose readings.

Parameters:

• df (pandas.DataFrame) – DataFrame containing the CGM values. The dataframe must contain ‘CGM’ and ‘Timestamp’ columns present in glucopy.Gframe.data.

• time_unit (str, default 'm' (minutes)) – The time unit for the x-axis. Can be ‘s (seconds)’, ‘m (minutes)’, or ‘h (hours)’.

• threshold (int | float, default 0) – The threshold value above which the AUC will be calculated.

• above (bool, default True) – If True, the AUC will be calculated above the threshold. If False, the AUC will be calculated below the threshold.

Returns:

auc – Area Under the Curve (AUC).

Return type:

float

Notes

This function is meant to be used by glucopy.Gframe.auc()