Gait patterns are a result of the complex kinematics that enable human two-legged locomotion, and they can reveal a lot about a person’s state and health. Analysing them is useful for researchers to get new insights into the course of diseases, and for physicians to track the progress after healing from injuries. When a person walks and is interfered with in any way, the resulting disturbance can show up and be found in the gait patterns. This paper describes an experimental setup for capturing gait patterns with a capacitive sensor floor, which can detect the time and position of foot contacts on the floor. With this setup, a dataset was recorded where 42 participants walked over a sensor floor in different modes, inter alia, normal pace, closed eyes, and dual-task. A recurrent neural network based on Long Short-Term Memory units was trained and evaluated for the classification task of recognising the walking mode solely from the floor sensor data. Furthermore, participants were asked to do the Unilateral Heel-Rise Test, and their gait was recorded before and after doing the test. Another neural network instance was trained to predict the number of repetitions participants were able to do on the test. As the results of the classification tasks turned out to be promising, the combination of this sensor floor and the recurrent neural network architecture seems like a good system for further investigation leading to applications in health and care.

Background/Aims The ability to climb stairs is an important prerequisite for activities of daily living and social participation in older adults, and is therefore an important part of rehabilitation. However, there is no consensus on how to measure stair-climbing ability. The aim of this study was to investigate the test–retest reliability of the measurement of stair-climbing speed (steps per second) as a parameter for functional ability in older adults. Methods A total of 57 participants who were in hospital and 56 participants who were community-dwelling and did not have any limitations in activities in daily living, all aged 60 years and over, ascended and descended a set of 13 stairs twice. The halfway point of the staircase was marked in order to split the time required for both the ascending and the descending actions. Additional measurements consisted of the Functional Reach Test, the Timed Up and Go Test, walking ability using the GAITRite walkway system and the isometric strength of four muscle groups of the lower extremities using a handheld dynamometer. Results Test–retest reliability of the first and second half of the stair-climbing for both ascending and descending showed excellent results for the group of hospitalised participants (intraclass correlation coefficient, [ICC] 0.87, 95% confidence interval [CI] 0.79–0.93 to 0.94, 95% CI 0.9 – 0.97 for comparison of first vs second half of stair climbing; ICC 0.9, 95% CI 0.83-0.94 to ICC 0.95, 95% CI 0.92–0.97 for comparing first vs second measurement)) and moderate to excellent results for the group of community-dwelling participants with no limitations (ICC 0.58, 95% CI 0.37–0.73 to ICC 0.76, 95% 95% CI 0.63-0.85 for comparison of first vs second half of stair climbing; ICC 0.82, 95% CI 0.71-0.89 to 0.92, 95% CI 0.87–0.95 for comparing first vs second measurement). As expected, hospitalised participants took significantly longer descending than ascending stairs (t(56)=6.98, P<0.001, d=0.93). A general and significant trend of increasing speed while descending could be observed in both groups (performing paired sample t-tests). Conclusions The results indicate that stair-climbing speed is not constant and that different patterns exist in older adults who have no limitations and in those who are hospitalised. The use of stair-climbing speed as an assessment tool should include both stair ascent and descent, because differences in these speeds seem to be indicators of stair-climbing ability.

No abstract