A wearable system for pre-impact fall detection.

Unique features of body segment kinematics in falls and activities of daily living (ADL) are applied to make automatic detection of a fall in its descending phase, prior to impact, possible. Fall-related injuries can thus be prevented or reduced by deploying fall impact reduction systems, such as an inflatable airbag for hip protection, before the impact. In this application, the authors propose the following hypothesis: "Thigh segments normally do not exceed a certain threshold angle to the side and forward directions in ADL, whereas this abnormal behavior occurs during a fall activity". Torso and thigh wearable inertial sensors (3D accelerometer and 2D gyroscope) are used and the whole system is based on a body area network (BAN) for the comfort of the wearer during a long term application. The hypothesis was validated in an experiment with 21 young healthy volunteers performing both normal ADL and fall activities. Results show that falls could be detected with an average lead-time of 700 ms before the impact occurs, with no false alarms (100% specificity), a sensitivity of 95.2%. This is the longest lead-time achieved so far in pre-impact fall detection.

Application of motion analysis system in pre-impact fall detection.

The purpose of this study is to investigate unique features of body segments in fall and activities of daily living (ADL) to make automatic detection of fall in its descending phase before the impact. Thus, fall-related injuries can be prevented or reduced by deploying feedback systems before the impact. In this study, the authors propose the following hypothesis: (1) thigh segment normally does not go beyond certain threshold angle to forward and sideways directions in ADL and (2) even if it does, the angular characteristics measured at torso and thigh differ from one another in ADL whereas in the case of fall, they become congruent. These two factors can be used to distinguish fall from ADL in its inception. Vicon 3-D motion analysis system was used in this study. High level of correlation between thigh and torso segments (corr > 0.99) was found for fall activities and low correlation coefficients (mean corr for lateral movements is 0.2338 and for sagittal movements is -0.665) were observed in ADL. By applying the hypothesis, all simulated falls could be detected with no false alarms and around 700ms lead-time before the impact was achieved in pre-impact fall detection. It is the longest lead-time obtained so far in pre-impact fall detection.

Distinguishing fall activities from normal activities by angular rate characteristics and high-speed camera characterization.

Distinguishing sideways and backward falls from normal activities of daily living using angular rate sensors (gyroscopes) was explored in this paper. Gyroscopes were secured on a shirt at the positions of sternum (S), front of the waist (FW) and right underarm (RU) to measure angular rate in lateral and sagittal planes of the body during falls and normal activities. Moreover, the motions of the fall incidents were captured by a high-speed camera at a frame rate of 250 frames per second (fps) to study the body configuration during fall. The high-speed camera and the sensor data capture system were activated simultaneously to synchronize the picture frame of high-speed camera and the sensor data. The threshold level for each sensor was set to distinguish fall activities from normal activities. Lead time of fall activities (time after threshold value is surpassed to the time when the hip hits the ground) and relative angle of body configuration (angle beta between the vertical line and the line from the center point of the foot or the center point between the two legs to that of the waist) at the threshold level were studied. For sideways falls, lead times of sensors at positions FW and S were about 200-220ms and 135-182ms, respectively. The lead time of the slippery backward fall (about 98ms) from the sensor at position RU was shorter than that of the sideways falls from the sensors at positions FW and S. The relative angle of body configuration at threshold level for sideways and backward falls were about 40-43 degrees for the sensor at position FW, about 43-52 degrees for the sensor at position S and about 54 degrees for the sensor at position RU, respectively. This is the first study that investigates fall dynamics in detection of fall before the person hits the ground using angular rate sensors (gyroscopes).

Classification of gait patterns in the time-frequency domain.

This paper describes the classification of gait patterns among descending stairs, ascending stairs and level walking activities using accelerometers arranged in antero-posterior and vertical direction on the shoulder of a garment. Gait patterns in continuous accelerometer records were classified in two steps. In the first step, direct spatial correlation of discrete dyadic wavelet coefficients was applied to separate the segments of gait patterns in the continuous accelerometer record. Compared to the reference system, averaged absolute error 0.387 s for ascending stairs and 0.404 s for descending stairs were achieved. The overall sensitivity and specificity of ascending stairs were 98.79% and 99.52%, and those of descending stairs were 97.35% and 99.62%. In the second step, powers of wavelet coefficients of 2 s time duration from separated segments of vertical and antero-posterior acceleration signals were used as features in classification. Our results proved a reliable technique of measuring gait patterns during physical activity.