Body sway measurement system

A body sway measurement system, comprising:

• a camera configured to generate time series output data for determining one or more parameters indicative of the body sway of a user;
• and
• a computing device including a data processor, the data processor including at least one hardware component, the data processor being operatively coupled to the camera, the data processor configured to receive the time series output data from the camera, and the data processor and/or a cloud server programmed to:
• process the time series output data to extract movement features including body displacement and/or sway patterns;
• extract displacement metrics from the time series output data collected from the camera focused on a stationary target remote from the user, the displacement metrics including a sway path displacement of the user over time starting from a sway path reference position, the sway path displacement of the user being determined by tracking a position of the stationary target in a plurality of frames of video data acquired by the camera, wherein a greater sway displacement from the sway path reference position is indicative of a lower stability of the user;
• and
• determine, using a trained neural network, the one or more parameters indicative of the body sway of the user based upon the displacement metrics extracted from the time series output data collected from the camera.

A body sway measurement system, comprising:

• an inertial measurement unit and/or camera configured to generate time series output data in a primary domain for determining one or more parameters indicative of the body sway of a user;
• and
• a mobile device including at least one of the inertial measurement unit and the camera so as to form a self-contained body sway measurement device, the mobile device further including a data processor, the data processor including at least one hardware component, the data processor being operatively coupled to the inertial measurement unit and/or camera, the data processor configured to receive the time series output data from the inertial measurement unit and/or camera, and the data processor and/or a cloud server programmed to:
• process the time series output data to extract movement features including body displacement and/or sway patterns while the mobile device is being worn or held by the user;
• extract secondary domain features from the time series output data collected from an accelerometer and/or gyroscope of the inertial measurement unit by applying one or more domain transform operations to accelerometer time series output data and/or gyroscope time series output data so as to identify dominant components corresponding to the sway patterns indicative of a fall risk of the user, and/or extract displacement metrics from the time series output data collected from the camera focused on a stationary target remote from the user, the displacement metrics including a sway path displacement of the user starting from a sway path reference position, wherein a greater sway displacement from the sway path reference position is indicative of a lower stability of the user;
• and
• determine, using a trained neural network, the one or more parameters indicative of the body sway of the user based upon the secondary domain features extracted from the time series output data collected from the inertial measurement unit, and/or based upon the displacement metrics extracted from the time series output data collected from the camera.