Comparison of wrist actimetry variables of paretic upper limb use in post stroke patients for ecological monitoring

FYI. What is wrist actimetry?

An actigraph is worn like a watch on the wrist of your non-dominant hand and measures activity through light and movement. Actigraphy data can be very helpful for assessing circadian rhythm disorders such as advanced or delayed sleep phase disorder and insomnia.

i can see almost no use for this, measuring lack of use does nothing to get survivors recovered. Deliver stroke protocols for recovery, that would help a hell of a lot more than useless measurements like this.

 

 

Comparison of wrist actimetry variables of paretic upper limb use in post stroke patients for ecological monitoring

• Gilles Dusfour,
• Denis Mottet,
• Makii Muthalib,
• Isabelle Laffont &
• Karima Bakhti 

Journal of NeuroEngineering and Rehabilitation volume 20, Article number: 52 (2023) Cite this article

• Metrics details

Abstract

Background

To date, many wrist actimetric variables dedicated to measuring the upper limbs (UL) in post-stroke patients have been developed but very few comparisons have been made between them. The objective of this study was to compare different actimetric variables of the ULs between a stroke and healthy population.

Methods

Accelerometers were worn continuously for a period of 7 days on both wrists of 19 post-stroke hemiparetic patients as well as 11 healthy subjects. Various wrist actimetry variables were calculated, including the Jerk ratio 50 (JR50, cumulative probability that the Jerk Ratio is between 1 and 2), absolute (FuncUse30) and relative (FuncUseRatio30) amounts of functional use of movements of the ULs with angular amplitude greater than 30°, and absolute (UH) and relative (UseHoursRatio) use hours.

Results

FuncUse30, FuncUseRatio30, UH, UseHoursRatio and JR50 of the paretic UL of stroke patients were significantly lower than in the non-dominant UL of healthy subjects. Comparing the ratio variables in stroke patients, FuncUseRatio30 was significantly lower than UseHoursRatio and JR50, suggesting a more clinically sensitive variable to monitor. In an exploratory analysis, FuncUseRatio tends to decrease with angular range of motion for stroke patients while it remains stable and close to 1 for healthy subjects. UseHoursRatio, FuncUseRatio30 and JR50 show linear correlation with Fugl-Meyer score (FM), with r² equal to 0.53, 0.35 and 0.21, respectively.

Conclusion

This study determined that the FuncUseRatio30 variable provides the most sensitive clinical biomarker of paretic UL use in post-stroke patients, and that FuncUseHours—angular range of motion relationship allows the identification of the UL behaviour of each patient. This ecological information on the level of functional use of the paretic UL can be used to improve follow-up and develop patient-specific therapy.

Background

Stroke is one of the leading causes of disability worldwide, with a global prevalence rate that has been increasing over the past 30 years [1]. Despite the accumulated research on rehabilitation of the paretic upper limb (UL) following a stroke, a large majority of patients continue to present non-use of paretic UL at the chronic stage which impacts their quality of daily life [2]. Only 5 to 20% of stroke survivors regain sufficient paretic UL function after 6 months [3], which leaves the majority of chronic post stroke patients unable to use their paretic UL in their daily life.

Current methods of quantifying movement of the UL rely primarily on clinical deficit scores such as the Fugl-Meyer (FM) test [4], or on more functional tests like the Wolf Motor Function Test (WMFT), Action Research Arm Test (ARAT) or questionnaires (Motor Activity Log—MAL). More recent work focused on the direct visual observation of stroke patients ULs by hospital practitioners in a clinical environment during 7 days [5]. This work found that the ratio of use activity between the paretic limb and the non-paretic limb is around 0.69 for stroke patients [5] whereas it is 0.95 for healthy subjects (non-dominant/dominant) [Bailey et al., 2014]. The human assessor method used by McLaren [5] has the advantage of identifying with certainty the periods of functional use of the UL as assessed directly by the clinician. However, the time and human resource costs of performing these measurements reduce its applicability to monitor multiple patients, and moreover, limiting observations in a clinical setting might not reflect real life of patients in a home environment.

Alternatively, a commonly used quantitative and objective technique to quantify functional UL movements relies on methods based on actimeters or gyroscopes [Bailey et al., 2014] positioned on the two wrists over a period of time ranging from 2 to 7 days. The functional UL movement results of Bailey's work [6, 7] are based on the calculation of activity counts directly from the acceleration signals originally developed by Uswatte [8]. Bailey derived other variables from the accelerometric measurements, such as use hours based on acceleration thresholds and median bilateral magnitude based on the magnitude of the accelerations measured at each wrist. The correlations of UL activity count with clinical scores such as the FM or the WMFT showed high variability between studies [6, 7]. Although Lang et al. [9] showed a strong correlation (r² = 0.62) between the use hours and the WMFT, a more recent study shows a weaker correlation between the median bilateral magnitude and the FM Score (r² = 0.32) or the WMFT score (r² = 0.34) [10]. Recently, Pan et al. [11] developed new accelerometric variables based on the Jerk, which is the derivative of acceleration. Pan et al. [11] showed that the Jerk ratio (JR) has a very high sensitivity to the amount of UL motion as well as a very high correlation with the median bilateral magnitude. Leuenberger et al. [12] extended the method by using inertial sensors (i.e., accelerometer and gyroscope) to separate functional vs. non-functional UL movements. A functional UL movement occurs when the forearm is oriented horizontally (± 30°), which is usually the case in UL manipulation activities. Leuenberger et al. [12] found excellent correlation of the functional use ratio (FuncUseRatio30) with the box and block (BB) test (r² = 0.9). Lum et al. [13] chose to synchronise accelerometers with video recordings of healthy and hemiparetic subjects performing activities of daily life. The video recordings were used both to accurately measure the amount of functional UL movement in the laboratory over a given period of time and to serve as a basis for labelling actimetric data as functional, non-functional and unknown movement. This labelling was then used to develop several machine learning algorithms to separate functional from non-functional UL movements. Although the activity counts showed low correlation with the video results (r² = 0.57), the machine learning algorithms showed excellent results (r² = 0.81).

The majority of studies on wrist actimetric monitoring of post-stroke patients monitor the quantity of UL movement and its bilateral ratio. While these indicators seem relevant for measuring imbalances between the UL motion, it remains difficult to draw conclusions about the functional imbalance and the physical capacities and limitations of the patients' UL in their ecological environments. Leuenberger's work provides a significant advance by identifying the amount of movement around the horizontal plane with an amplitude of ± 30°. For the first time, it is possible to easily discriminate a "functional" amount of movement in an ecological environment. However, Leuemberger's work required inertial sensors that are relatively expensive and have little energy autonomy. Finally, more recent studies based on artificial intelligence algorithms show promising results in identifying functional motion but require a significant amount of time for manual classification of motion. Moreover, these algorithms do not yet distinguish between movements of different amplitudes. This makes it difficult to identify the physical capabilities and limitations of post-stroke subjects in their ecological environments while it is necessary to turn accelerometer data into clinical meaningful data [14].

In this study we recorded 3D acceleration at each wrist, over a period of 7 days, in the volunteers’ home (ecological) environment. We then adapted and compared different accelerometric variables (UseHours, UseHoursRatio, FuncUse30, FuncUseRatio30 and JR50) between a population of 19 stroke patients and 11 healthy subjects to determine the actimetric variable that has the greatest sensitivity to stroke hemiparesis-related upper limb deficits in order to guide clinical decision making.

More at link.