Rapid Dynamic Naturalistic Monitoring of Bradykinesia in Parkinson’s Disease Using a Wrist-Worn Accelerometer

Abstract

Motor fluctuations in Parkinson’s disease are characterized by unpredictability in the timing and duration of dopaminergic therapeutic benefits on symptoms, including bradykinesia and rigidity. These fluctuations significantly impair the quality of life of many Parkinson’s patients. However, current clinical evaluation tools are not designed for the continuous, naturalistic (real-world) symptom monitoring needed to optimize clinical therapy to treat fluctuations. Although commercially available wearable motor monitoring, used over multiple days, can augment neurological decision making, the feasibility of rapid and dynamic detection of motor fluctuations is unclear. So far, applied wearable monitoring algorithms are trained on group data. In this study, we investigated the influence of individual model training on short timescale classification of naturalistic bradykinesia fluctuations in Parkinson’s patients using a single-wrist accelerometer. As part of the Parkinson@Home study protocol, 20 Parkinson patients were recorded with bilateral wrist accelerometers for a one hour OFF medication session and a one hour ON medication session during unconstrained activities in their own homes. Kinematic metrics were extracted from the accelerometer data from the bodyside with the largest unilateral bradykinesia fluctuations across medication states. The kinematic accelerometer features were compared over the 1 h duration of recording, and medication-state classification analyses were performed on 1 min segments of data. Then, we analyzed the influence of individual versus group model training, data window length, and total number of training patients included in group model training, on classification. Statistically significant areas under the curves (AUCs) for medication induced bradykinesia fluctuation classification were seen in 85% of the Parkinson patients at the single minute timescale using the group models. Individually trained models performed at the same level as the group trained models (mean AUC both 0.70, standard deviation respectively 0.18 and 0.10) despite the small individual training dataset. AUCs of the group models improved as the length of the feature windows was increased to 300 s, and with additional training patient datasets. We were able to show that medication-induced fluctuations in bradykinesia can be classified using wrist-worn accelerometry at the time scale of a single minute. Rapid, naturalistic Parkinson motor monitoring has the clinical potential to evaluate dynamic symptomatic and therapeutic fluctuations and help tailor treatments on a fast timescale.

Assessment of upper limb movement disorders using wearable sensors during functional tasks: a systematic review

“…Current overview can support clinicians and researchers to select the most sensitive pathology-dependent sensor features and measurement methodologies for detection and quantification of upper limb movement disorders and for the objective evaluations of treatment effects. Accelerometer PD [63,89,96], dystonia [140] , ataxia [134], MS [127]…”

“…Teeth brushing was included in four studies [52,68,77,84] as well as putting clothes on/off [47,48,77,87]. In stroke, the Wolf Motor Function test or parts of this clinical scale were included four times [115,[118][119][120] and four studies measured activities in an unrestricted home environment [63,89,134,139]. Combing hair was included in three studies [52,77,87] as well as typing and folding laundry [76,77,86] and forwards and sideways reaching [12,114,116].…”

“…For the studies where both acceleration and angular velocity signals were collected, both mean and standard deviation (STD) were often calculated [62,78,86,87,90,118,129,139], as well as rootmean-square (RMS) values [11,86,89,129,139]. Additionally, mean and RMS or STD of acceleration and angular velocity separately were used in studies were one of the signals was available [29,50,52,54,57,63,64,66,71,74,82,83,85,89,96,104,105,108,110,111,115,119,120,[132][133][134]137], as well as median angular velocity in one study [82]. Maximal linear velocity was additionally often used as key feature, mostly by integration of the acceleration signal [51,54,115].…”

“…For signal smoothness, RMS of jerk was often used as a straightforward measure in PD [118] and stroke [115,116], as well as a jerk metric for which multiple definitions were given, mostly RMS jerk normalized over time/peak velocity or mean jerk [12,15,76,89,90,115,123,139,141].…”

Wearables for Movement Analysis in Healthcare

“…In clinical applications, wearable sensors were used in several pathological states, such as stroke [ 4 , 5 , 6 ], obese [ 7 , 8 ], and elderly [ 9 ] patients; patients with lower limb amputation [ 10 ]; patients with Parkinson’s disease [ 4 , 11 , 12 ]; and patients hospitalized for knee joint rehabilitation [ 13 ]. In particular, wearable systems were used both to quantify the functional limitations of the patients, during several movements (gait [ 5 , 7 , 8 , 9 ], upper limb [ 6 , 11 ], time up and go test [ 10 ], and unconstraint activities at home [ 12 ]) and to evaluate their accuracy and precision in comparison with the gold standard [ 4 ]. These papers support the clinical usability of wearable technology for clinical movement assessment.…”