Validation of Freezing-of-Gait Monitoring Using Smartphone.

BACKGROUND: Freezing of gait (FOG) is a commonly observed motor symptom for patients with Parkinson's disease (PD). The symptoms of FOG include reduced step lengths or motor blocks, even with an evident intention of walking. FOG should be monitored carefully because it not only lowers the patient's quality of life, but also significantly increases the risk of injury. INTRODUCTION: In previous studies, patients had to wear several sensors on the body and another computing device was needed to run the FOG detection algorithm. Moreover, the features used in the algorithm were based on low-level and hand-crafted features. In this study, we propose a FOG detection system based on a smartphone, which can be placed in the patient's daily wear, with a novel convolutional neural network (CNN). METHODS: The walking data of 32 PD patients were collected from the accelerometer and gyroscope embedded in the smartphone, located in the trouser pocket. The motion signals measured by the sensors were converted into the frequency domain and stacked into a 2D image for the CNN input. A specialized CNN model for FOG detection was determined through a validation process. RESULTS: We compared our performances with the results acquired by the previously reported settings. The proposed architecture discriminated the freezing events from the normal activities with an average sensitivity of 93.8% and a specificity of 90.1%. CONCLUSIONS: Using our methodology, the precise and continuous monitoring of freezing events with unconstrained sensing can assist patients in managing their chronic disease in daily life effectively.

Wrist sensor-based tremor severity quantification in Parkinson's disease using convolutional neural network.

Tremor is a commonly observed symptom in patients of Parkinson's disease (PD), and accurate measurement of tremor severity is essential in prescribing appropriate treatment to relieve its symptoms. We propose a tremor assessment system based on the use of a convolutional neural network (CNN) to differentiate the severity of symptoms as measured in data collected from a wearable device. Tremor signals were recorded from 92 PD patients using a custom-developed device (SNUMAP) equipped with an accelerometer and gyroscope mounted on a wrist module. Neurologists assessed the tremor symptoms on the Unified Parkinson's Disease Rating Scale (UPDRS) from simultaneously recorded video footages. The measured data were transformed into the frequency domain and used to construct a two-dimensional image for training the network, and the CNN model was trained by convolving tremor signal images with kernels. The proposed CNN architecture was compared to previously studied machine learning algorithms and found to outperform them (accuracy = 0.85, linear weighted kappa = 0.85). More precise monitoring of PD tremor symptoms in daily life could be possible using our proposed method.

Tremor frequency characteristics in Parkinson's disease under resting-state and stress-state conditions.

Tremor characteristics-amplitude and frequency components-are primary quantitative clinical factors for diagnosis and monitoring of tremors. Few studies have investigated how different patient's conditions affect tremor frequency characteristics in Parkinson's disease (PD). Here, we analyzed tremor characteristics under resting-state and stress-state conditions. Tremor was recorded using an accelerometer on the finger, under resting-state and stress-state (calculation task) conditions, during rest tremor and postural tremor. The changes of peak power, peak frequency, mean frequency, and distribution of power spectral density (PSD) of tremor were evaluated across conditions. Patients whose tremors were considered more than "mild" were selected, for both rest (n=67) and postural (n=25) tremor. Stress resulted in both greater peak powers and higher peak frequencies for rest tremor (p<0.001), but not for postural tremor. Notably, peak frequencies were concentrated around 5 Hz under stress-state condition. The distributions of PSD of tremor were symmetrical, regardless of conditions. Tremor is more evident and typical tremor characteristics, namely a lower frequency as amplitude increases, are different in stressful condition. Patient's conditions directly affect neural oscillations related to tremor frequencies. Therefore, tremor characteristics in PD should be systematically standardized across patient's conditions such as attention and stress levels.

Unconstrained detection of freezing of Gait in Parkinson's disease patients using smartphone.

Freezing of gait (FOG) is a common motor impairment to suffer an inability to walk, experienced by Parkinson's disease (PD) patients. FOG interferes with daily activities and increases fall risk, which can cause severe health problems. We propose a novel smartphone-based system to detect FOG symptoms in an unconstrained way. The feasibility of single device to sense gait characteristic was tested on the various body positions such as ankle, trouser pocket, waist and chest pocket. Using measured data from accelerometer and gyroscope in the smartphone, machine learning algorithm was applied to classify freezing episodes from normal walking. The performance of AdaBoost.M1 classifier showed the best sensitivity of 86% at the waist, 84% and 81% in the trouser pocket and at the ankle respectively, which is comparable to the results of previous studies.

Clinicians' Tendencies to Under-Rate Parkinsonian Tremors in the Less Affected Hand.

The standard assessment method for tremor severity in Parkinson's disease is visual observation by neurologists using clinical rating scales. This is, therefore, a subjective rating that is dependent on clinical expertise. The objective of this study was to report clinicians' tendencies to under-rate Parkinsonian tremors in the less affected hand. This was observed through objective tremor measurement with accelerometers. Tremor amplitudes were measured objectively using tri-axis-accelerometers for both hands simultaneously in 53 patients with Parkinson's disease during resting and postural tremors. The videotaped tremor was rated by neurologists using clinical rating scales. The tremor measured by accelerometer was compared with clinical ratings. Neurologists tended to under-rate the less affected hand in resting tremor when the contralateral hand had severe tremor in Session I. The participating neurologists corrected this tendency in Session II after being informed of it. The under-rating tendency was then repeated by other uninformed neurologists in Session III. Kappa statistics showed high inter-rater agreements and high agreements between estimated scores derived from the accelerometer signals and the mean Clinical Tremor Rating Scale evaluated in every session. Therefore, clinicians need to be aware of this under-rating tendency in visual inspection of the less affected hand in order to make accurate tremor severity assessments.

Adaptive windowing for gait phase discrimination in Parkinsonian gait using 3-axis acceleration signals.

In order to robustly analyze the gait of Parkinson's disease (PD) patients, a new gait phase discrimination method was developed for analyzing the three-axis accelerations of the ankle during walking. The magnitude of acceleration was compared with the lowpass-filtered signal of itself and pseudo foot-flat phases were determined. Four narrow windows were made sequentially and adaptively from the pseudo foot-flat phases. Each window contained a characteristic peak that discriminated the gait phases. From these windows, the initial contact (IC) point and end contact (EC) point were determined by finding the maximal point in the proximal-distal acceleration. Seven healthy individuals and 17 PD patients were subjected to a walking test on level ground for a distance of 6.5 m with the wearable activity monitoring system (W-AMS). Foot pressure and movement images were simultaneously recorded as references. The ICs and ECs detected by the proposed algorithm were compared with the manually marked events in the foot pressure signals. In healthy subjects, all the ICs and ECs were correctly detected. In the PD group, the detection accuracy was 97.6% for the ICs and 99.4% for the ECs. Based on these results, this novel method holds promise for use in monitoring temporal gait parameters continuously in PD patients, which will subsequently allow for the evaluation of motor fluctuations in PD patients.

Classification of Parkinson gait and normal gait using Spatial-Temporal Image of Plantar pressure.

The purpose of this paper is the classification of Spatial-Temporal Image of Plantar pressure (STIP) among normal step and the patients step of Parkinson disease. For this, we created a new image data, STIP, that have information of the change of plantar pressure during heel to toe motion (i.e., contain spatial and temporal information for plantar pressure). To get STIP, the walking of 21 patients with Parkinson disease and 17 age-matched healthy subjects were recorded and analyzed using in-shoe dynamic pressure measuring system with comfort walking. For feature extraction of gait, we applied Principal component analysis (PCA) to STIP and calculated weights of STIP on each principal components. Then, we build hard margin Support Vector Machine (SVM) classifier for gait recognition and test of generalization performance using normalized weights on PCs of STIP. SVM result indicated an overall accuracy of 91.73% by the RBF(Radial Basis Function) kernel function. These results demonstrate considerable potential in applying SVMs in gait classification for many applications.