Skip to main content

Advertisement

Table 9 Comparison of the results on transitions to gait models with varying parameters

From: Developing a system that can automatically detect health changes using transfer times of older adults

Changing parameters Scenarioa FPRb DRc ARLd
μ and σ V SU 0.20±0.07 100 % 7.70±3.34
  V US 0.17±0.07 100 % 10.95±3.59
  V SUS 0.20±0.08 100 % 10.40±4.50
  V USU 0.20±0.08 100 % 9.05±3.33
μ only V S T2 0.20±0.09 100 % 7.85±2.94
  V U T1 0.16±0.06 100 % 9.55±3.78
  V S T2S 0.23±0.04 100 % 8.05±3.41
  V U T1U 0.21±0.07 97.50 % 9.97±3.59
σ only V S T1 0.13±0.08 50 % 14.50±6.96
  V U T2 0.09±0.07 25 % 7.20±9.47
  V S T1S 0.15±0.06 35 % 11.57±9.07
  V U T2U 0.11±0.09 37.50 % 12.87±8.25
  1. This change in gait model parameters is defined as either a change in both μ, σ or either μ or σ (all transitions have a length of four weeks)
  2. Notes
  3. aSimulation scenarios as described in Table 4
  4. When two trends are present in the data
  5. an average DR and ARL is calculated
  6. bAverage number of False Alerts per week
  7. cDetection Rate
  8. dAverage Run Length