Kentaro Yagi1, Yuta Sugiura2, Kunihiro Hasegawa3, Hideo Saito4. 1. Department of information and Computer Science, Keio University, 3-14-1 Hiyoshi Kohoku-ku, Yokohama, 223-8522, Japan. Electronic address: yagi@hvrl.ics.keio.ac.jp. 2. Department of information and Computer Science, Keio University, 3-14-1 Hiyoshi Kohoku-ku, Yokohama, 223-8522, Japan. Electronic address: sugiura@keio.jp. 3. Department of information and Computer Science, Keio University, 3-14-1 Hiyoshi Kohoku-ku, Yokohama, 223-8522, Japan. Electronic address: hiro@hvrl.ics.keio.ac.jp. 4. Department of information and Computer Science, Keio University, 3-14-1 Hiyoshi Kohoku-ku, Yokohama, 223-8522, Japan. Electronic address: hs@keio.jp.
Abstract
BACKGROUND: An individual's gait is a key factor for consideration in evaluating their overall health. Several medical studies have demonstrated the correlation between gait and incidence rate of diseases, mortality, and risk of fall. However, gait is only occasionally evaluated during medical visits, which may delay the detection of health problems. METHODS: In this paper, we propose a gait measurement system that is suitable for use at home. Our method requires only a single RGB camera, whereas other visionary sensor-based methods require depth sensors or multiple RGB cameras. In addition, the setup for the measurement is easy. What the user has to do is only putting a single camera in a room and choosing four location known points on the floor. Our method can measure step positions and step timings, and therefore, other important parameters such as stride length, step width, walking speed, and cadence may also be captured. The individual's gait is captured by the camera, and therefore the user is not required to wear any devices. RESULTS: In the experiment described herein, we demonstrate our method's accuracy by comparing it with the motion capture system. The results indicate that our method can measure walking speed with an error of 3.62 cm/s from the side view, and which is too small a change to be clinically meaningful.
BACKGROUND: An individual's gait is a key factor for consideration in evaluating their overall health. Several medical studies have demonstrated the correlation between gait and incidence rate of diseases, mortality, and risk of fall. However, gait is only occasionally evaluated during medical visits, which may delay the detection of health problems. METHODS: In this paper, we propose a gait measurement system that is suitable for use at home. Our method requires only a single RGB camera, whereas other visionary sensor-based methods require depth sensors or multiple RGB cameras. In addition, the setup for the measurement is easy. What the user has to do is only putting a single camera in a room and choosing four location known points on the floor. Our method can measure step positions and step timings, and therefore, other important parameters such as stride length, step width, walking speed, and cadence may also be captured. The individual's gait is captured by the camera, and therefore the user is not required to wear any devices. RESULTS: In the experiment described herein, we demonstrate our method's accuracy by comparing it with the motion capture system. The results indicate that our method can measure walking speed with an error of 3.62 cm/s from the side view, and which is too small a change to be clinically meaningful.
Authors: Yohanna MejiaCruz; Jean Franco; Garret Hainline; Stacy Fritz; Zhaoshuo Jiang; Juan M Caicedo; Benjamin Davis; Victor Hirth Journal: Curr Geriatr Rep Date: 2021-01-20