Table 3 Estimated marginal means for each meal detection algorithm.

From: Performance of continuous glucose monitoring-based meal detection algorithms in young healthy adults

Algorithm

Sensitivity (%)

FP/day

Δt (min)

MDAKölle-CGM13

85.2 [79.1–89.8]

0.38 [0.26–0.56]

43.8 [38.9–48.8]

MDADassau−2of33

74.9 [67.2–81.3]

0.59 [0.43–0.80]

37.4 [32.3–42.4]

MDADassau−3of43

48.9 [40.3–57.6]

0.12 [0.06–0.24]

37.2 [31.9–42.4]

MDAFaccioli16

66.4 [57.9–73.9]

1.37 [1.12–1.69]

39.3 [34.2–44.3]

MDAHarvey15

73.0 [65.0–79.6]

0.26 [0.17–0.41]

37.6 [32.6–42.7]

MDAKölle-Ra13

80.2 [73.2–85.7]

0.33 [0.22–0.49]

41.7 [36.8–46.7]

MDAPopp12

85.7 [79.6–90.1]

1.26 [1.01–1.57]

59.1 [54.2–64.1]

MDASamadi18

92.0 [87.4–95.0]

2.39 [2.02–2.82]

59.1 [54.1–64.0]

MDATurksoy14

79.7 [72.7–85.3]

0.22 [0.13–0.36]

40.1 [35.1–45.1]

  1. Data are mean [95% CI].
  2. MDA meal detection algorithm, TP true positives, FP false positives, FN false negatives, FP/day false positives per day, Δt detection time.
Back to article page