Toward Smart Footwear to Track Frailty Phenotypes—Using Propulsion Performance to Determine Frailty

Abstract

Frailty assessment is dependent on the availability of trained personnel and it is currently limited to clinic and supervised setting. The growing aging population has made it necessary to find phenotypes of frailty that can be measured in an unsupervised setting for translational application in continuous, remote, and in-place monitoring during daily living activity, such as walking. We analyzed gait performance of 161 older adults using a shin-worn inertial sensor to investigate the feasibility of developing a foot-worn sensor to assess frailty. Sensor-derived gait parameters were extracted and modeled to distinguish different frailty stages, including non-frail, pre-frail, and frail, as determined by Fried Criteria. An artificial neural network model was implemented to evaluate the accuracy of an algorithm using a proposed set of gait parameters in predicting frailty stages. Changes in discriminating power was compared between sensor data extracted from the left and right shin sensor. The aim was to investigate the feasibility of developing a foot-worn sensor to assess frailty. The results yielded a highly accurate model in predicting frailty stages, irrespective of sensor location. The independent predictors of frailty stages were propulsion duration and acceleration, heel-off and toe-off speed, mid stance and mid swing speed, and speed norm. The proposed model enables discriminating different frailty stages with area under curve ranging between 83.2–95.8%. Furthermore, results from the neural network suggest the potential of developing a single-shin worn sensor that would be ideal for unsupervised application and footwear integration for continuous monitoring during walking.

Assessment of frailty: a survey of quantitative and clinical methods

“…High frequency components during a standing tandem test have also proved to differentiate [42], 5m Walk Test [54] Pre-Frail, Frail and Robust [42], Classified as Frail, Frail with Mild Cognitive and Robust [54] 3 Gait Symmetry A. Martinez-Ramirez et al [42] Tri-Axial Inertial Orientation [42] 3 m W a l k T e s t [ 42] Pre-frail, Frail and Robust [42] 4 Gait Variability A. Martinez-Ramirez et al [54] Tri-Axial Inertial Sensor Lumbar Spine (L3) Acceleration Signal (Vertical Direction Only) [54] 5m Walk Test [54] Classified as Frail, Frail with Mild Cognitive and Robust [54] 5 Signal Root Mean Square (RMS) Value A.Martinez-Ramirez et al [42] Tri-Axial Inertial Orientation [42] 3 m W a l k T e s t [ 42] Pre-Frail, Frail and Robust [42] 6 Stride Length Walk of 4.5m [26], TUG [19] Pre-Classified Using the Fried Phenotype [26],Correlation of TFI, TUG and Gait Parameters [19] 9 Swing Time E. Gianaria et al [19] Kinect Sensor [19] T U G [ 19] Correlation of TFI, TUG and Gait Parameters [19] 10 Stride Velocity M. Schwenk et al [26] 5 inertial sensor unit shank, thighs and lower back [26] Walk of 4.5m [26] Pre-classified using the Fried Phenotype [26] 11 Cadence N.A. Capela et al [45] Smart Phone [45] 2 -6 m i n W a l k T e s t [ 45] 12 Dual Task Gait Patterns I n e r t i a l S e n s o r [ 43] Walking Test [43] Pre-Classified using the Fried Phenotype [43] healthy subjects from pre-frail and frail subjects. However, frequency pattern...…”

“…The location of inertial sensor is critical to extract meaningful gait parameters. The location of inertial sensors at the lower extremities, e.g the shin or the ankle position, can generate gait patterns in the vertical direction and reveal more meaningful kinematic parameters as gait activity depends on the movements of lower extremities [43].…”

“…Rahemi and colleagues [43] developed a frailty model using six sensor-derived gait parameters of Toe-off speed, mid-Swing speed, mid-stance speed, propulsion duration, propulsion acceleration and speed norm. The parameters were derived from the angular velocity signal of sensors attached on the right and the left lower shin of the subject.…”

Smart Socks and In-Shoe Systems: State-of-the-Art for Two Popular Technologies for Foot Motion Analysis, Sports, and Medical Applications

“…The gyroscope overcomes the disadvantages of accelerometers, which are sensitive to the sensor location, and also of pressure sensors, which need to be calibrated regularly. The proposed algorithm could be implemented, as the authors stated, in Sensoria smart socks, in smart shoes, and others, in which the gyroscope sensors can be integrated [40].…”

Wearable Sensors to Monitor, Enable Feedback, and Measure Outcomes of Activity and Practice

“…In addition, the standard machine learning pipeline for sensor-based movement assessment consisting of inertial signal pre-processing, feature extraction, and classifier training is increasingly being compared to deep learning frameworks for movement classification accuracy [20]. For example, disease or disability phenotypes were discerned by using sensorderived gait parameters to distinguish among non-frail, prefrail, and frail persons [21]. The key parameters obtained with shank sensors and derived by an artificial neural network analysis were propulsion duration and acceleration, heel-off and toe-off speed, mid-stance and mid-swing speed, and speed norms.…”