Background: To quantify pharmacological effects on tremor in patients with essential tremor (ET) or Parkinson's Disease (PD), laboratory-grade accelerometers have previously been used. Over the last years, consumer products such as smartphones and smartwatches have been increasingly applied to measure tremor in an easy way. However, it is unknown how the technical performance of these consumer product accelerometers (CPAs) compares to laboratory-grade accelerometers (LGA). This study was performed to compare the technical performance of CPAs with LGA to measure tremor in patients with Parkinson's Disease (PD) and essential tremor (ET). Methods: In ten patients with PD and ten with ET, tremor peak frequency and corresponding amplitude were measured with 7 different CPAs (Apple iPhone 7, Apple iPod Touch 5, Apple watch 2, Huawei Nexus 6P, Huawei watch, mbientlabMetaWear (MW) watch, mbientlab MW clip) and compared to a LGA (Biometrics ACL300) in resting and extended arm position. Results: Both in PD and ET patients, the peak frequency of CPAs did not significantly differ from the LGA in terms of limits of agreement. For the amplitude at peak frequency, only the iPhone and MW watch performed comparable to the LGA in ET patients, while in PD patients all methods performed comparable except for the iPod Touch and Huawei Nexus. Amplitude was higher when measured with distally-located CPAs (Clip, iPhone, iPod) compared with proximally-located CPAs (all watches). The variability between subjects was higher than within subjects for frequency (25.1% vs. 13.4%) and amplitude measurement (331% vs. 53.6%). Resting arm position resulted in lower intra-individual variability for frequency and amplitude (13.4 and 53.5%) compared to extended arm position (17.8 and 58.1%).

IntroductionChronic low back pain (CLBP) is a major public health problem. Reliably measuring the effects of chronic pain on movement and activity, and any changes due to treatment, is a healthcare challenge. A recently published paper demonstrated that a novel digital therapeutic (DTxP) was efficacious in reducing fear of movement and increasing the quality of life of adult patients with moderate to severe CLBP. In this paper, we report a study of how data from wearable devices collected in this study could be used as a digital measure for use in studies of chronic low back pain.MethodsMovement, electrodermal recording, general activity and clinical assessment data were collected in a clinical trial of a novel digital therapeutic intervention (DTxP) by using the sensors in commercial Garmin Vivosmart 4, Empatica Embrace2 and Oculus Quest wearables. Wearable data were collected during and between the study interventions (frequent treatment sessions of DTxP). Data were analyzed using exploratory statistical analysis.ResultsA pattern of increased longitudinal velocity in the movement data collected with right-hand, left-hand, and head sensors was observed in the study population. Correlations were observed with the changes in clinical scales (Tampa Scale of Kinesiophobia, EQ5D Overall health VAS, and EQ5D QoL score). The strongest correlation was observed with the increased velocity of head and right-hand sensors (Spearman correlation with increasing head sensor velocity and Tampa Scale of Kinesiophobia −0.45, Overall health VAS +0.67 and EQ5D QoL score −0.66). The sample size limited interpretation of electrodermal and general activity data.Discussion/ConclusionWe found a novel digital signal for use in monitoring the efficacy of a digital therapeutics (DTxP) in adults with CLBP. We discuss the potential use of such movement based digital markers as surrogate or additional endpoints in studies of chronic musculoskeletal pain.Clinical Trial Registrationhttps://clinicaltrials.gov/ct2/show/NCT04225884?cond=NCT04225884&draw=2&rank=1, identifier: NCT04225884.

In the quantification of symptoms of Parkinson’s disease (PD), healthcare professional assessments, patient reported outcomes (PRO), and medical device grade wearables are currently used. Recently, also commercially available smartphones and wearable devices have been actively researched in the detection of PD symptoms. The continuous, longitudinal, and automated detection of motor and especially non-motor symptoms with these devices is still a challenge that requires more research. The data collected from everyday life can be noisy and frequently contains artefacts, and novel detection methods and algorithms are therefore needed. 42 PD patients and 23 control subjects were monitored with Garmin Vivosmart 4 wearable device and asked to fill a symptom and medication diary with a mobile application, at home, for about four weeks. Subsequent analyses are based on continuous accelerometer data from the device. Accelerometer data from the Levodopa Response Study (MJFFd) were reanalyzed, with symptoms quantified with linear spectral models trained on expert evaluations present in the data. Variational autoencoders (VAE) were trained on both our study accelerometer data and on MJFFd to detect movement states (e.g., walking, standing). A total of 7590 self-reported symptoms were recorded during the study. 88.9% (32/36) of PD patients, 80.0% (4/5) of DBS PD patients and 95.5% (21/22) of control subjects reported that using the wearable device was very easy or easy. Recording a symptom at the time of the event was assessed as very easy or easy by 70.1% (29/41) of subjects with PD. Aggregated spectrograms of the collected accelerometer data show relative attenuation of low (<5Hz) frequencies in patients. Similar spectral patterns also separate symptom periods from immediately adjacent non-symptomatic periods. Discriminative power of linear models to separate symptoms from adjacent periods is weak, but aggregates show partial separability of patients vs. controls. The analysis reveals differential symptom detectability across movement tasks, motivating the third part of the study. VAEs trained on either dataset produced embedding from which movement states in MJFFd could be predicted. A VAE model was able to detect the movement states. Thus, a pre-detection of these states with a VAE from accelerometer data with good S/N ratio, and subsequent quantification of PD symptoms is a feasible strategy. The usability of the data collection method is important to enable the collection of self-reported symptom data by PD patients. Finally, the usability of the data collection method is important to enable the collection of self-reported symptom data by PD patients.

Chronic low back pain (CLBP) is a globally common musculoskeletal problem. Measuring the sensation of pain and the effect of a treatment has always been a challenge for healthcare. Here, we study how the movement data, collected while using a virtual reality (VR) program, could be used as an objective measurement in patients with CLBP. A specific data collection method based on VR was developed and used with CLBP patients and healthy volunteers. We demonstrate that the movement data in VR can be used to classify individuals in these two groups with a high accuracy by using logistic regression. The most discriminative features are the duration of the movements and the total variation of movement velocity. Furthermore, we show that hidden Markov models can divide movement data into meaningful segments, which creates possibilities for defining even more detailed features, with potential to improve accuracy, when larger datasets become available in the future.