Quantitative evaluation of gait characteristics was conducted in patients with lumbar spinal stenosis (LSS). Our method utilized a miniature three-dimensionalinertialsensor attached on the upper trunk of a subject during a clinical walk test. Six LSS subjects (age 66±3 years) participated in this study. Measurements before surgery and at three months after surgery were compared. Gait capability was examined by the six-minute walk test, with the patient walking back and forth a 30-meter straight corridor in a hospital. First, gait parameters were estimated from the sensor output, including step counts, cadence and mean step lengths. Furthermore three-dimensional rotation angles of the upper trunk were calculated to illustrate postural sway features. Time-variable characteristics that express pain or intermittent claudication were evaluated by the path length and range of postural sway trajectory during walking. Experimental results showed remarkable changes in posturalperturbation over time before surgery. Walk distance and duration were improved after surgery. Analysis of postural sway revealed reduced anterior-posterior sway range after surgery, with maintenance of upper body steadiness and smooth walking. The proposed inertia sensor used in combination with the conventionalcl inicalwal king test may be usefulfor evaluating the therapeutic effects in LSS patients in terms of gait kinematic characterization.
Electroencephalography (EEG) is a traditional method used in sleep research, but remains inappropriate for routine measurements due to the need for physical restraint and complex attachment of electrodes. For routine monitoring, measurement during sleep in an unencumbered and unrestrained state is essential. The present study focused on body movements during sleep. We extracted the characteristics of body movements during sleep by varying the body sites measured. We measured movements of the following parts of the body during sleep hours:head, trunk, limbs, and whole-body. A NapVIEW was used to detect movements of the head and an Actigraph to detect trunk and limb movements. Additionally, an infrared sensor (NaPiOn) suspended from the ceiling was used to detect whole-body movement. For comparing different sensing sites and different methods of measurement, “body movement timing” was defined as a measure of the rate of temporal concordance with body movement. Results of body movement timing showed high concordance between head and whole-body movements. Significant differences (p<0.001, p<0.01) were observed between limb movement and whole-body movement. Occurrence rates of four kinds of body movement in seven states of sleep were also analyzed. Significant differences were observed in the arousal state during sleep and a state of sleep level 6-8. These results suggest that the arousal state during sleep and a state of sleep level 6-8 can be estimated using body movements.