Figure 2
From: Passive longitudinal weight and cardiopulmonary monitoring in the home bed
Non-contact adherence-independent longitudinal weight monitoring. (a) Measurement of individual (black) and summed (purple) sensor loads during movement of a 25 lbs weight from positions A through E (middle to the four corners of the bed). (b) Cartoon illustrating a sleeping subject redistributing total body weight during episodic movements in bed. Individual (colored) and summed (black) sensor loads are shown while a subject is awake in bed using a laptop with frequent redistributions of load (left) and during sleep with only episodic movements separated by long periods of lying still (right). (c) 3 weeks of the longitudinal sum of scales used to derive daily weights. Inset shows a single day of a person getting into and out of bed. (d) Correlation plot of BedScales versus commercial floor scale weights (R2 = 0.99, n = 162). (e) Histogram of measurement error, expressed as a percent of measured weight, comparing BedScales versus commercial floor scale (error mean -0.057% and standard deviation 0.21%). (f) Bland–Altman comparing BedScales versus commercial floor scale weights. (g) Comparison of daily BedScales weight measurements compared to two commercial bathroom scales. Pairwise error mean (0.12%) and standard deviation (0.40%) are shown.
