Using Smart Socks to Detect Step-count at Slow Walking Speeds in Healthy Adults

Abstract

We examined the accuracy of Smart Socks - a device that measures foot pressure during gait for detecting step-count across various walking speeds. Thirty-six participants (17 men; 19 women) wore Smart Socks (Sock), a pedometer (Pedometer), and a smartphone with a commercially available Phone Application (Phone) pedometer to measure step-count during 3-min of treadmill or over-ground walking at 1.3, 2.2, 3.0, 3.8, and 4.7 km/h. Steps were compared to a gold-standard tally-counter (Count), collected by independent assessors. All devices (Sock, Pedometer, and Phone) underestimated step-count when compared to Count at 1.3 km/h (p<0.05); however, Sock (27±18%) demonstrated a lower percent error compared to Phone (40±28%) and Pedometer (98±5%) (both p<0.01). At 2.2 km/h, Sock was not different compared to Count (Sock: 213±39; Count: 229±24 steps, p=0.25); however, both Phone (271±55 steps) and Pedometer (169±166 steps) were different compared to Count (p<0.05). At 3.0 km/h, both Sock (258±30 steps) and Pedometer (254±45 steps) were similar to Count (267±22 steps) (p>0.05); however, Phone (291±28 steps) overestimated step-count (p<0.01). All devices (Sock, Pedometer, and Phone) were similar to Count at 3.8, and 4.7 km/h (p>0.05). These findings demonstrate that Smart Socks are more accurate than pedometers used in the present study for detecting step-count during treadmill or over-ground ambulation at slower walking speeds.