概要

Home-Based Monitor for Gait and Activity Analysis

Published: August 08, 2019
doi:

概要

This innovative device uses magneto-inertial sensors to permit gait and activity analysis in uncontrolled environments. Currently in the qualification process as an outcome measure in the European Medical Agency, one of the applications will be to serve as a clinical endpoint in clinical trials in neuromuscular diseases.

Abstract

Current outcomes in neuromuscular disorder clinical trials include motor function scales, timed tests, and strength measures performed by trained clinical evaluators. These measures are slightly subjective and are performed during a visit to a clinic or hospital and constitute therefore a point assessment. Point assessments can be influenced by daily patient condition or factors such as fatigue, motivation, and intercurrent illness. To enable home-based monitoring of gait and activity, a wearable magneto-inertial sensor (WMIS) has been developed. This device is a movement monitor composed of two very light watch-like sensors and a docking station. Each sensor contains a tri-axial accelerometer, gyroscope, magnetometer, and a barometer that record linear acceleration, angular velocity, the magnetic field of the movement in all directions, and barometric altitude, respectively. The sensors can be worn on the wrist, ankle, or wheelchair to record the subject’s movements during the day. The docking station enables data uploading and recharging of sensor batteries during the night. Data are analyzed using proprietary algorithms to compute parameters representative of the type and intensity of the performed movement. This WMIS can record a set of digital biomarkers, including cumulative variables, such as total number of meters walked, and descriptive gait variables, such as the percentage of the most rapid or longest stride that represents the top performance of patient over a predefined period of time.

Introduction

A number of potential therapies are in development for treatment of genetic neuromuscular diseases. These diseases include Duchenne muscular dystrophy (DMD) and spinal muscular atrophy (SMA) type 3. Subjects with these diseases present initially with proximal lower limb weakness that leads to progressive difficulties in ambulation. The final step in translational research is the demonstration of efficacy of a potential treatment or approach in a clinical trial. Specific, quantifiable, objective, and reliable measures are required. The importance of such measures was recently emphasized by the failure of the phase IIb ataluren trial1 and the phase III Biomarin trial2. One of the likely explanations for these failures was the variability and the nonlinear evolution of the primary outcome measure of these trials, the 6-minute walk test3 (6 MWT). Increasing reliability and sensitivity to the change of outcome measures and the understanding of the factors leading to their variation could contribute to decrease the number of trial failures related to the main outcome measures.

One of the limitations of the current outcomes is the subjectivity of the assessment. To further increase the objectivity of assessments, Heberer et al.4 showed that through a marker set and the use of a gait analysis software, there was a significant increase in stride length in patients treated with steroids compared with the naïve group. Hip joint kinetics are early markers of proximal weakness in patients with DMD and are responsive to change with steroid intervention, which is the only available treatment for these patients. Gait laboratories are, however, only available in large clinics. Furthermore, laboratory evaluations are point assessments, and a patient’s condition may greatly vary on a day to day basis due to factors such as fatigue, motivation, and intercurrent illness.

The use of continuous and home-based measurement should achieve both a more objective and a more globally representative assessment. In other fields of neurology, for instance Parkinson5 or multiple sclerosis6, several studies have assessed the feasibility, reliability, and consistency with other measures of different sensors including accelerometers with or without gyrometers or magnetometers, yet none of these devices is currently a gold standard for evaluation of patients during clinical trials. In the field of neuromuscular diseases, there is currently no validated method for continuous home monitoring of patients. In recent years, through a close collaboration between clinicians and engineers, the Institute of Myology in Paris has developed several devices for upper limb assessment to precisely evaluate upper limb strength and function7,8,9. A wearable magneto-inertial sensor (WMIS; i.e., ActiMyo) has been developed in collaboration with a company specialized in navigation systems. Initially a monitoring device dedicated to non-ambulant subjects with neuromuscular disorders such as DMD and SMA10,11, the same device has now been used to monitor ambulant patients in two different configurations: sensors on both ankles or one sensor at the wrist and the other one at the ankle. The configuration for a non-ambulant population is composed of a sensor at the wheelchair and the other one at the wrist.

This WMIS is able to precisely capture and quantify all movements of the limb on which it is placed. The measuring principle is based on the use of microelectromechanical system (MEMS) inertial sensors and magnetometers operated through magneto-inertial equations. Dedicated algorithms allow precise qualification and quantification of patients’ movements in a non-controlled environment. 

The overall goal of the method is to provide identification and quantification of any movement produced by a patient over a pre-defined period of time, and to integrate these measures into disease-specific outcome measures representative of the patient’s condition over a period of time.

To effectively assess ambulant and non-ambulant patients with movement disorders at home, the device must be provided to the patient by a trained evaluator who is responsible for making sure that the instructions have been understood. An investigator and a patient manual are provided with the device. This WMIS is currently being used as an exploratory outcome measure in a number of clinical trials for neuromuscular and neurologic diseases (NCT03351270, NCT02780492, NCT01385917, NCT03039686, NCT03368742, NCT02500381). Specific procedures adapted to the pathology and/or to the clinical trial design have been developed.

Protocol

Any use of the device must be carried out in accordance with the rules established by the reference protocol, validated by the ethics committee and the national regulatory agencies of the country. The use of the device and the various elements attached to it must be done within the intended use described in the patient's manual.

NOTE: To be eligible to use of the WMIS, patient must be over 5 years old, be able to understand and follow the usage rules, provide informed consent, be affiliate or beneficiary of a social security scheme, and be able to comply with all protocol requirements. There are no specific exclusion criteria.

1. Preparing for the participant’s visit at the clinical center

  1. Check the suitcase contents: (1) the docking station to plug the sensors during the night for data uploading onto a USB key and recharging of batteries, (2) the power cord divided into two pieces to connect the docking station to a power supply, (3) the Ethernet cable to permit interface with a router, (4) the two sensors to permit daily activity recording, (5) bands for attachment of the sensors depending on the chosen configuration and ambulatory status of the participant (ankle-ankle: two ankle armbands with two stickers to distinguish the wearing side; ankle-wrist: one ankle armband and one bracelet; wrist-wheelchair: one bracelet and one wheelchair pocket), (6) one participant manual and one investigator manual, (7) one task reminder, (8) a screwdriver to enable replacement of the USB key, and (9) two blank USB keys.
    NOTE: If an internet connection is available data transfer occurs automatically
  2. Print and prepare the assignment form to record the assignment of a device to a participant. This will enable data reconciliation with the subject ID.

2. Training of the subject during the first visit

  1. Positioning of this WMIS
    1. Based on the data that the researchers are seeking as well as the patient’s ambulatory status, use different configurations for the placement of the sensors.
      1. For ambulant patients, fix the two sensors either on a wrist and an ankle for upper and lower limb activity recording or on both ankles for only lower limbs activity.
      2. For non-ambulant patients, fix one sensor on the wrist and the other one on the wheelchair.
  2. Explain positioning of the sensors to the participant.
    1. For ambulant participants, fix the two sensors either on a wrist and an ankle or on both ankles.
      1. For the wrist-ankle configuration, place one sensor on the wrist of the dominant hand using the provided bracelet so that the waves are pointing toward the fingers. Place the second sensor on the ankle, on the same side as the wrist sensor, above the external malleolus with the waves pointing in the forward direction.
        NOTE: The sensor must be placed on top of the wrist.
      2. For the ankle-ankle configuration, place a sensor on each ankle, above the external malleolus with the waves pointing in the forward direction.
        NOTE: Stickers should be placed on the sensors to indicate the wearing side.
    2. For non-ambulant participants, fix one sensor on the wrist and the other one on the wheelchair.
      1. For the wrist-wheelchair configuration, place one sensor on the wrist of the participant’s dominant hand using the provided bracelet so that the waves are pointing toward the fingers. Place the second sensor in the bag provided. Attach it on a safe place to the wheelchair.
        NOTE: Independently of the used configuration, do not switch sensors around. The sensors should fit snuggly, but not too tightly to the wrist and/or ankle to prevent them from spinning around.
  3. Explain daily routine for sensors’ use to the participant.
    1. Evening routine
      1. Plug the docking station into the power supply. Attach the docking station to the router if an internet connection is available. Insert the sensors into the docking station.
      2. Localize the two light-emitting diodes (LEDs) on the docking station which indicate the status of the sensors. Once plugged into a power source, be sure the station beeps and the diodes become orange to indicate that the sensor batteries are charging, and that data is being downloaded from the sensors to the USB drive.
        NOTE: If the LEDS are still blinking after 5 min, restart the procedure from beginning. If the issue persists, contact the clinical site team.
    2. Morning routine
      1. Verify that the LEDs are green, indicating that the sensors batteries are fully charged, and that data has been cleared from the sensors’ memory. Remove the sensors from the docking station. Wear the sensors in the configuration demonstrated by the evaluator.
        NOTE: If LED on one or both sensors is orange after two consecutive days, contact the clinical center.
    3. Daytime routine
      1. Wear the sensors the entire day and place the sensors back on the docking station at the end of the day.
        NOTE: Remove the sensors during activities involving water, special medical examinations (e.g., magnetic resonance imaging [MRI], CT-scan, X-ray) or any activity that could damage them, and keep them in a safe place on a firm surface. Resume wearing the sensors after the activity.
    4. At the end of the recording period, tidy up all the device items in the suitcase and bring back the device to the clinical center.
      NOTE: Encourage the participant to engage in normal daily activities wearing as much as possible the sensors.
  4. Complete a dedicated assignment form.

3. Data collection and analysis

  1. Data collection
    NOTE: The sensors record signals continuously for up to 16 h and store the information in an internal memory (Movie 1). The docking station enables the downloading of data stored in the sensors at the end of each day and charging of the batteries during the night. Data downloaded to the docking station is stored on a USB drive that can only be accessed by the evaluators.
    1. A standard 64 GB USB drive can hold up to 3 months of daily recording information (approximately 16 h/day). Provide higher or lower capacity USB drives to adjust as closely as possible to the constraints of the protocols.
    2. If the docking station is not connected to internet, have the evaluator remove the USB drive from the docking station (with the specific screwdriver contained in the suitcase) and replace it with a blank one at the end of the recording period. The USB drive should be sent to the support team for analysis.
      NOTE: If the docking station is connected to the internet, data are uploaded to cloud storage. Thus, there is no need to change the USB drive, as all data are automatically deleted from the USB drive once the files are uploaded into the cloud.
  2. Data analysis
    1. At selected time points during the study, extract data from the cloud storage and analyze data using a dedicated algorithm. Adjust the analysis periods and the monitoring reports based on the clinical study.

Representative Results

Data presented here were acquired during clinical trials approved by the ethics committee and the French Regulatory Agency. All patient representatives signed an informed consent.

This WMIS was first used in a clinical study setting in 2012 for controlled and home-based monitoring of upper-limb movements in non-ambulant DMD patients (NCT01611597), which demonstrated the autonomy and feasibility of device use10. Variables, such as norm of angular velocity, the ratio of the vertical component of the acceleration to the overall acceleration, the elevation rate, and the computed power, were identified to clinically characterize the upper limb activity of patients in a controlled environment (Table 1). In a second step, these variables were correlated with the efficacy of patients during a standardized and validated task, which also allowed testing of device reliability10. A more complete validation of variables relevant to non-ambulatory subjects is ongoing.

The validation process in ambulant patients is much more advanced. Recently, the committee for medicinal products for human use (CHMP) wrapped up the public consultation for the qualification of the 95th centile stride velocity (95CSV) as a validated secondary outcome in ambulant DMD patients and the final adoption of qualification opinion is pending. This WMIS allows the identification and measure (height, length, velocity) of each single stride (Figure 1) over a long period of time and enables analysis of the distribution of all captured strides, which allows calculation of centile of stride speed and length. The 95CSV appeared to be the most sensitive variable to changes in ambulant DMD patients. The 95CSV provides a sensitive outcome measure that allows a small number of patients per group in a controlled trial. Several other parameters measureable by this WMIS are less sensitive to change than 95CSV but are more closely related to quality of life such as distance walked per hour and the number of falls per hour (Table 1).

The precision of the gait trajectory was tested initially in controls, using an optokinetic system that confirmed excellent agreement between trajectories measured by optokinetic system and magnetoinertial sensor12. In patients, we assessed the agreement between the distances measured by this WMIS during a 6 MWT to the distance measured by physiotherapists. Figure 2 illustrates the ankle trajectory and orientation reconstructed from our WMIS measurements during one lap of a 6 MWT. For this study, data were obtained from 23 ambulant DMD patients (NCT02780492) during 31 6 MWT (some patients performed a second test 6 months later). Results are displayed in Figure 3. The difference between the distance measured by our WMIS and the reference 6 MWT (after correction for the length of the turn around the cones, see Figure 2) was within 5%.

Figure 1
Figure 1: Representation of the stride length by the black line which is the computed trajectory of the ankle position during walk reconstructed by this WMIS. Please click here to view a larger version of this figure.

Figure 2
Figure 2: Representation of the trajectory during one lap of a 6 MWT reconstructed by this WMIS. Please click here to view a larger version of this figure.

Figure 3
Figure 3: The 6 MWT distance calculated using this WMIS (y axis) versus the distance measured by the physiotherapist during the same test (x axis) for 31 6 MWT performed by 23 patients. Please click here to view a larger version of this figure.

Movie 1
Movie 1: Reconstruction of the ankle trajectory during walking by using this WMIS. Please click here to view this video. (Right-click to download.)

Walking parameters Upper limb parameters
Percentage of time spent walking (%) Percentage of time involving upper limb activity (%)
Walked distance per hour (m/h) Median angular velocity of the wrist (°/s)
Median stride speed (m/s) 95th angular velocity of the wrist (°/s)
95CSV (m/s) Median vertical acceleration (g)
Median stride length (m) 95th vertical acceleration (g)
95th percentile of the stride length (m) Median power (W/kg)
Number of falls per hour (count/h) 95th power (W/kg)

Table 1: List of parameters.

Discussion

In the past decade, a number of different systems have been developed, such as an activity monitor (Table of Materials [IV]), which uses accelerometric sensors to monitor activities of daily life for energy expenditure quantification13. A triaxial accelerometer (Table of Materials [V]) was used by Tanaka et al.14 to monitor activity of preschool children. Lau et al.15 showed through the combination of a dual-accelerometer (Table of Materials [VI]) and a gyroscope (Table of Materials [VII]) that gait spatiotemporal characteristics can be precisely determined by the use of inertial sensors. Zijlstra et al.16 analyzed trunk and 289 compensatory movements of the pelvis occurred during gait using a body-fixed sensor. Most devices used for gait analyses are three-axis accelerometers17,18,19 (Table of Materials [VIII, IX, X]). In addition to accelerometers, our WMIS includes three gyroscopes, a magnetometer, and a barometer. Inertial sensors constitute an interesting technology for evaluation of the motricity of neuromuscular patients due to the difficulty of assessing their reduced movements and their abnormal gaits. Our WMIS can be used to evaluate even the most severely impaired patients, and unlike scales and other tools, provides objective and reliable data. Indeed, the precise measure of strides cannot be achieved with simple wrist worn inertial sensors but requires highly stable and precisely calibrated inertial sensors sampled at a high frequency and placed on the lower limb in order to compute stride trajectory. The same is true of other exploratory outcomes, in ambulant patients, like stair climbing or falls.

We demonstrated that precise estimate of foot trajectory in ambulant DMD is feasible by using our WMIS, which can be worn during daily life. The validated 6 MWT3 and the North Star Ambulatory Assessment (NSAA)20 have been correlated with our device’s variables describing spontaneous walk during two weeks recording and are sensitive to change in the DMD population over a 6 month period.

Upper limb activity discrimination during the use of the wheelchair can be challenging. Accelerometry permits the monitoring of physical activity either directly related to wheelchair21 or upper limb movements during wheelchair movements22 (Table of Materials [XI]). By placing a sensor on the wheelchair and using data from the three-axial accelerometer, gyroscope, and barometer, our WMIS can distinguish upper limb activity from wheelchair and caregiver movements and precisely quantify even very weak movements.

The small size of our sensors is in line with recommendations by Ciuti et al.23 who noted that the miniaturization of sensors allows new applications in various fields. Improvements in technology now permit easy use, access and promote home-based monitoring of patients through data transmission systems integrated into smart devices, such as smartphones24,25,26. However, sensors in smartphones are not calibrated to be used as a clinical outcome measure, and the variability over time is not controlled. Thus, smartphones or tablets can be used for patient-reported outcome measures (Table of Materials [XII]) but not as a device to measure movements precisely and continuously.

Several limitations are associated with our device and its protocol. One of the main challenges with continuous home recording is that patient compliance tends to decrease over time. In order to tackle this problem, we had to find the minimal period of time with the lowest variability of the defined variables that can be considered clinically significant for a patient. This was defined as 180 hours of recording, which corresponds to two weeks27.

A critical step of the protocol is the training. The patient must be trained to use the device, and it is crucial that the evaluator providing the device and the training has first been trained adequately. The best way to optimize the training of the clinical team and the patients to ensure consistent use of the device is by face-to-face training.

Our WMIS will find applications outside the neuromuscular field, for example in evaluation of patients with multiple sclerosis or Parkinson disease, in which other less sensitive and reliable devices have already been tested28,29. Qualification processes taking into account measure reliability, variability, confounding factors, minimally meaningful differences, and sensitivity to change must be performed for these subjects as they have been carried out in ambulant patients with DMD described here.

開示

The authors have nothing to disclose.

Acknowledgements

The authors thank Anne-Gaëlle Le Moing, Amélie Moreaux, and Eric Dorveaux for their contribution to the development of this wearable magneto-inertial sensor and Jackie Wyatt for the review.

Materials

ActiMyo Sensors Sysnav SF-000080 Wearable magneto-inertal sensors attached to the patient for movment recording
Helen Hayes marker set Vicon NA Whole body jumpsuit with predefined Vicon's spots
OrthoTrak (Motion Analysis, Santa Rosa, CA, USA) Motion Lab Systems Gait analysis software
ActiGraph ActiGraph Corp GTM1 Activity monitor, used by researchers to capture and record continuous, high resolution physical activity and sleep/wake information
ActivTracer GMS LTD GMS Co. Ltd Japan AC-301A Triaxial accelerometer
ADXL202E dual-accelerometer Analog Devices ADXL212AEZ High precision, low power, complete dual axis accelerometer with signal conditioned, duty cycle modulated outputs, all on a single monolithic IC.
ENC-03J gyroscope Murata Electronics ENC-03J Vibration Sensors
DynaPort MiniMod MCROBERTS Small and light case containing a tri-axial accelerometer, a rechargeable battery, an USB connection, and raw data storage on a MicroSD card
MM-2860 Sunhayato Sunhayato MM-2860 3-axis accelerometer
MicroStone MA3-10Ac MA3-04AC Microstone Co. Acceleration sensors
RT3 Activity monitor Abledata NA Triaxial accelerometer
Aparito aparito NA Wearables and disease specific mobile apps to deliver patient monitoring outside of the hospital; Elin Davies, Aparito: https://www.aparito.com/
Docking station Sysnav SF-000118
Sensor Sysnav SF-000080
Bracelet
(black/grey L)
(black/grey S) (black/yellow L) (black/yellow S)
Sysnav ZZ-000093 ZZ-000094 ZZ-000247 ZZ-000248
Patient manual Sysnav FD-000086
Ethernet cable (2 m max.) Sysnav IC-000458
Power cable
(EU)
(UK)
(US)
Sysnav ZE-000440 ZE-000441 ZE-000442
Power supply unit Sysnav ZE-000443
Ankle strap Sysnav ZZ-000462
Small bag Sysnav ZZ-000033

参考文献

  1. McDonald, C. M., et al. Ataluren in patients with nonsense mutation Duchenne muscular dystrophy (ACT DMD): a multicentre, randomised, double-blind, placebo-controlled, phase 3 trial. The Lancet. 390 (10101), 1489-1498 (2017).
  2. Aartsma-Rus, A., et al. Development of Exon Skipping Therapies for Duchenne Muscular Dystrophy: A Critical Review and a Perspective on the Outstanding Issues. Nucleic Acid Therapeutics. 27 (5), 251-259 (2017).
  3. McDonald, C. M., et al. The 6-minute walk test as a new outcome measure in Duchenne muscular dystrophy. Muscle & Nerve. 41 (4), 500-510 (2010).
  4. Heberer, K., et al. Hip kinetics during gait are clinically meaningful outcomes in young boys with Duchenne muscular dystrophy. Gait & Posture. 48, 159-164 (2016).
  5. Pan, W., et al. Actigraphy monitoring of symptoms in patients with Parkinson’s disease. Physiology & Behavior. 119, 156-160 (2013).
  6. Supratak, A., et al. Remote Monitoring in the Home Validates Clinical Gait Measures for Multiple Sclerosis. Frontiers in Neurology. 9, 561 (2018).
  7. Seferian, A. M., et al. Upper Limb Strength and Function Changes during a One-Year Follow-Up in Non-Ambulant Patients with Duchenne Muscular Dystrophy: An Observational Multicenter Trial. PloS One. 10 (2), e0113999 (2015).
  8. Seferian, A. M., et al. Upper Limb Evaluation and One-Year Follow Up of Non-Ambulant Patients with Spinal Muscular Atrophy: An Observational Multicenter Trial. PLoS One. 10 (4), e0121799 (2015).
  9. Servais, L., et al. Innovative methods to assess upper limb strength and function in non-ambulant Duchenne patients. Neuromuscular Disorders. 23 (2), 139-148 (2013).
  10. Le Moing, A. G., et al. A Movement Monitor Based on Magneto-Inertial Sensors for Non-Ambulant Patients with Duchenne Muscular Dystrophy: A Pilot Study in Controlled Environment. PLoS One. 11 (6), e0156696 (2016).
  11. Chabanon, A., et al. Prospective and longitudinal natural history study of patients with Type 2 and 3 spinal muscular atrophy: Baseline data NatHis-SMA study. PLoS One. 13 (7), e0201004 (2018).
  12. Valentini, R., Martinelli, B., Mezzarobba, S., De Michiel, A., Toffano, M. Optokinetic analysis of gait cycle during walking with 1cm- and 2cm-high heel lifts. The Foot. 19 (1), 44-49 (2009).
  13. Rothney, M. P., Brychta, R. J., Meade, N. N., Chen, K. Y., Buchowski, M. S. Validation of the ActiGraph Two-Regression Model for Predicting Energy Expenditure. Medicine & Science in Sports & Exercise. 42 (9), 1785-1792 (2010).
  14. Tanaka, C., Tanaka, S. Daily Physical Activity in Japanese Preschool Children Evaluated by Triaxial Accelerometry: The Relationship between Period of Engagement in Moderate-to-Vigorous Physical Activity and Daily Step Counts. Journal of Physiological Anthropology. 28 (6), 283-288 (2009).
  15. Lau, H., Tong, K. The reliability of using accelerometer and gyroscope for gait event identification on persons with dropped foot. Gait & Posture. 27 (2), 248-257 (2008).
  16. Zijlstra, A., Goosen, J. H. M., Verheyen, C. C. P. M., Zijlstra, W. A body-fixed-sensor based analysis of compensatory trunk movements during unconstrained walking. Gait & Posture. 27 (1), 164-167 (2008).
  17. Brandes, M., Zijlstra, W., Heikens, S., van Lummel, R., Rosenbaum, D. Accelerometry based assessment of gait parameters in children. Gait & Posture. 24 (4), 482-486 (2006).
  18. Liu, K., Liu, T., Shibata, K., Inoue, Y., Zheng, R. Novel approach to ambulatory assessment of human segmental orientation on a wearable sensor system. Journal of Biomechanics. 42 (16), 2747-2752 (2009).
  19. Itoh, T., Morioka, I. Developing Leg-Motion Measurement System Using Acceleration Sensors. Electrical Engineering in Japan. 204 (1), 59-66 (2018).
  20. Scott, E., et al. Development of a functional assessment scale for ambulatory boys with Duchenne muscular dystrophy. Physiotherapy Research International. 17 (2), 101-109 (2012).
  21. Sonenblum, S. E., Sprigle, S., Caspall, J., Lopez, R. Validation of an accelerometer-based method to measure the use of manual wheelchairs. Medical Engineering & Physics. 34 (6), 781-786 (2012).
  22. Hiremath, S. V., Ding, D. Regression equations for RT3 activity monitors to estimate energy expenditure in manual wheelchair users. 2011 Annual International Conference of the IEEE Engineering in Medicine and Biology Society. , 7348-7351 (2011).
  23. Ciuti, G., Ricotti, L., Menciassi, A., Dario, P. M. E. M. S. Sensor Technologies for Human Centred Applications in Healthcare, Physical Activities, Safety and Environmental Sensing: A Review on Research Activities in Italy. Sensors. 15 (3), 6441-6468 (2015).
  24. del Rosario, M., Redmond, S., Lovell, N. Tracking the Evolution of Smartphone Sensing for Monitoring Human Movement. Sensors. 15 (8), 18901-18933 (2015).
  25. Shoaib, M., Bosch, S., Incel, O., Scholten, H., Havinga, P. Fusion of Smartphone Motion Sensors for Physical Activity Recognition. Sensors. 14 (6), 10146-10176 (2014).
  26. Miao, F., Cheng, Y., He, Y., He, Q., Li, Y. A Wearable Context-Aware ECG Monitoring System Integrated with Built-in Kinematic Sensors of the Smartphone. Sensors. 15 (5), 11465-11484 (2015).
  27. EMA. . Draft qualification opinion on stride velocity 95th centile as a secondary endpoint in Duchenne Muscular Dystrophy measured by a valid and suitable wearable device*. , (2018).
  28. Coulter, E. H., et al. Validity of the activPAL3 activity monitor in people moderately affected by Multiple Sclerosis. Medical Engineering & Physics. 45, 78-82 (2017).
  29. Thorp, J. E., Adamczyk, P. G., Ploeg, H. L., Pickett, K. A. Monitoring Motor Symptoms during Activities of Daily Living in Individuals with Parkinson’s Disease. Frontiers Neurology. 9, 1036 (2018).

Play Video

記事を引用
Lilien, C., Gasnier, E., Gidaro, T., Seferian, A., Grelet, M., Vissière, D., Servais, L. Home-Based Monitor for Gait and Activity Analysis. J. Vis. Exp. (150), e59668, doi:10.3791/59668 (2019).

View Video