Real-time gait analysis with accelerometer-based smart shoes.

In this paper, we present the evaluation of a new smart shoe capable of performing gait analysis in real time. The system is exclusively based on accelerometers which minimizes the power consumption. The estimated parameters are activity class (rest/walk/run), step cadence, ground contact time, foot impact (zone, strength, and balance), forward distance, and speed. The different parameters have been validated with a customized database of 26 subjects on a treadmill and video data labeled manually. Key measures for running analysis such as the cadence is retrieved with a maximum error of 2%, and the ground contact time with an average error of 3.25%. The classification of the foot impact zone achieves a precision between 72% and 91% depending of the running style. The presented algorithm has been licensed to ICON Health & Fitness Inc. for their line of wearables under the brand iFit.

An autonomous medical monitoring system: Validation on arrhythmia detection.

In this paper, we present a generic platform for autonomous medical monitoring and diagnostics. We validated the platform in the context of arrhythmia detection with publicly available databases. The big advantage of this platform is its capacity to deal with various types of physiological signals. Many pre-processing steps are performed to bring the input information into a uniform state that will be explored by a machine learning algorithm. Since this block plays a crucial role in the entire processing pipeline, three different methods were evaluated for detection and classification of anomalies. The results presented in this work are validated on cardiac beats, where the highest accuracy was obtained on the classification of normal beats (94%). On the other hand, atrial fibrillation and premature ventricular contraction beats were classified with an accuracy of 78%.

Real-time monitoring of swimming performance.

This article presents the performance results of a novel algorithm for swimming analysis in real-time within a low-power wrist-worn device. The estimated parameters are: lap count, stroke count, time in lap, total swimming time, pace/speed per lap, total swam distance, and swimming efficiency (SWOLF). In addition, several swimming styles are automatically detected. Results were obtained using a database composed of 13 different swimmers spanning 646 laps and 858.78 min of total swam time. The final precision achieved in lap detection ranges between 99.7% and 100%, and the classification of the different swimming styles reached a sensitivity and specificity above 98%. We demonstrate that a swimmers performance can be fully analyzed with the smart bracelet containing the novel algorithm. The presented algorithm has been licensed to ICON Health & Fitness Inc. for their line of wearables under the brand iFit.

Accurate walking and running speed estimation using wrist inertial data.

In this work, we present an accelerometry-based device for robust running speed estimation integrated into a watch-like device. The estimation is based on inertial data processing, which consists in applying a leg-and-arm dynamic motion model to 3D accelerometer signals. This motion model requires a calibration procedure that can be done either on a known distance or on a constant speed period. The protocol includes walking and running speeds between 1.8km/h and 19.8km/h. Preliminary results based on eleven subjects are characterized by unbiased estimations with 2(nd) and 3(rd) quartiles of the relative error dispersion in the interval ±5%. These results are comparable to accuracies obtained with classical foot pod devices.

Physical activity profiling: Activity-specific step counting and energy expenditure models using 3D wrist acceleration.

In this paper, we present the evaluation of a new physical activity profiling system embedded in a wrist-located device. We propose a step counting and an energy expenditure (EE) method, and evaluate their accuracy against gold standard references. To this end, we used an actimetry sensor on the waist and an indirect calorimetry monitoring device on a population of 13 subjects to obtain step count and metabolic equivalent task (kcal/kg/h) referenced values. The subjects followed a protocol that spanned a given set of activities (lying, standing, walking, running) at a wide range of intensities. The performance of the EE model was characterized by a root-mean-square error (RMSE) of 1.22±0.34kcal/min, and step-count model at regular walking/running speeds by 0.71±0.06step/10sec.

Validation of a wrist monitor for accurate estimation of RR intervals during sleep.

While the incidence of sleep disorders is continuously increasing in western societies, there is a clear demand for technologies to asses sleep-related parameters in ambulatory scenarios. The present study introduces a novel concept of accurate sensor to measure RR intervals via the analysis of photo-plethysmographic signals recorded at the wrist. In a cohort of 26 subjects undergoing full night polysomnography, the wrist device provided RR interval estimates in agreement with RR intervals as measured from standard electrocardiographic time series. The study showed an overall agreement between both approaches of 0.05 ± 18 ms. The novel wrist sensor opens the door towards a new generation of comfortable and easy-to-use sleep monitors.