Abstract
[Purpose] To examine optimal parts of the body to attach activity meters to measure hemiplegic patients’ numbers of steps. [Subjects and Methods] A 3-axis accelerometer-incorporated activity meter was attached to the bilateral upper and lower limbs and trunk of 23 patients with post-stroke hemiplegia and 20 healthy individuals to compare relative errors between their measured and actual numbers of steps. [Results] On comparing the hemiplegic and healthy groups, relative errors were higher among the former in the non-paretic upper limb, trunk, and non-paretic lower limb. In the healthy group, relative errors were higher in the upper limb compared with trunk and lower limb. In the hemiplegic group, relative errors were higher in the non-paretic upper limb compared with paretic upper limb, paretic lower limb, and non-paretic lower limb. [Conclusion] Leading to higher relative errors, non-paretic upper limbs may be inappropriate to attach activity meters. In contrast, paretic lower limbs may minimize errors, as the values obtained when attaching the meter to them were as low as those in the case of healthy lower limbs.
