Rainer Gloeckl1, Thomas Damisch2, Jos Prinzen3, Rob van Lummel3, Evert Pengel4, Ursula Schoenheit-Kenn2, Klaus Kenn2. 1. Schoen Klinik Berchtesgadener Land, Department of Respiratory Medicine & Pulmonary Rehabilitation, Schoenau am Koenigssee, Germany. Electronic address: rgloeckl@schoen-kliniken.de. 2. Schoen Klinik Berchtesgadener Land, Department of Respiratory Medicine & Pulmonary Rehabilitation, Schoenau am Koenigssee, Germany. 3. McRoberts, The Hague, The Netherlands. 4. The Hague University of Applied Sciences, Faculty of Technology, Innovation & Society, The Netherlands.
Abstract
INTRODUCTION: Objective of this study was to validate an activity monitor (DynaPort MoveMonitor [MM], McRoberts, The Hague, The Netherlands) against night-vision video analysis during sleep. METHODS: Twenty patients (65 ± 11 years old, mean body-mass-index: 27 ± 6 kg/m(2)) with different chronic lung diseases were recruited to participate in this validation study. Patients performed a polysomnography measurement during one single night while wearing the MM. The activity monitor data of the MM were then validated against the analysis of the night-vision video by an independent investigator. In total, four different lying positions (supine, left, right and prone), sitting upright, out of bed as well as large, medium, small and sitting transitions were classified. RESULTS: A mean duration of 7.6 ± 0.9 h per night of video and MM classification was available for analysis. In total, 702 different postures were registered on the video from which 678 postures (96.6%) were detected correctly by the MM compared to the video classification. These results yielded a total degree of sensitivity of 93.9% and specificity 94.9% in detecting postures during the night. In total, 682 transitions (394 small, 189 medium, 15 large and 84 sitting transitions) were detected of which 482 were also detected by the MM. The MM detected 70% of the transitions correctly (51.0% small, 97.4% medium, 100% large and 97.6% sitting transitions). CONCLUSION: The MM is an activity monitor showing a high degree of sensitivity and specificity to detect different nocturnal postures as well as medium and large sized transitions in patients with chronic respiratory disorders.
INTRODUCTION: Objective of this study was to validate an activity monitor (DynaPort MoveMonitor [MM], McRoberts, The Hague, The Netherlands) against night-vision video analysis during sleep. METHODS: Twenty patients (65 ± 11 years old, mean body-mass-index: 27 ± 6 kg/m(2)) with different chronic lung diseases were recruited to participate in this validation study. Patients performed a polysomnography measurement during one single night while wearing the MM. The activity monitor data of the MM were then validated against the analysis of the night-vision video by an independent investigator. In total, four different lying positions (supine, left, right and prone), sitting upright, out of bed as well as large, medium, small and sitting transitions were classified. RESULTS: A mean duration of 7.6 ± 0.9 h per night of video and MM classification was available for analysis. In total, 702 different postures were registered on the video from which 678 postures (96.6%) were detected correctly by the MM compared to the video classification. These results yielded a total degree of sensitivity of 93.9% and specificity 94.9% in detecting postures during the night. In total, 682 transitions (394 small, 189 medium, 15 large and 84 sitting transitions) were detected of which 482 were also detected by the MM. The MM detected 70% of the transitions correctly (51.0% small, 97.4% medium, 100% large and 97.6% sitting transitions). CONCLUSION: The MM is an activity monitor showing a high degree of sensitivity and specificity to detect different nocturnal postures as well as medium and large sized transitions in patients with chronic respiratory disorders.
Authors: Vincent Theodoor van Hees; S Sabia; S E Jones; A R Wood; K N Anderson; M Kivimäki; T M Frayling; A I Pack; M Bucan; M I Trenell; Diego R Mazzotti; P R Gehrman; B A Singh-Manoux; M N Weedon Journal: Sci Rep Date: 2018-08-28 Impact factor: 4.379