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.

Reconstruction of 12-lead ECG Using a Single-patch Device.

OBJECTIVES: The aim of this study is to develop an optimal electrode system in the form of a small and wearable single-patch ECG monitoring device that allows for the faithful reconstruction of the standard 12-lead ECG. METHODS: The optimized universal electrode positions on the chest and the personalized transformation matrix were determined using linear regression as well as artificial neural networks (ANNs). A total of 24 combinations of 4 neighboring electrodes on 35 channels were evaluated on 19 subjects. Moreover, we analyzed combinations of three electrodes within the four-electrode combination with the best performance. RESULTS: The mean correlation coefficients were all higher than 0.95 in the case of the ANN method for the combinations of four neighboring electrodes. The reconstructions obtained using the three and four sensing electrodes showed no significant differences. The reconstructed 12-lead ECG obtained using the ANN method is better than that using the MLR method. Therefore, three sensing electrodes and one ground electrode (forming a square) placed below the clavicle on the left were determined to be suitable for ensuring good reconstruction performance. CONCLUSIONS: Since the interelectrode distance was determined to be 5 cm, the suggested approach can be implemented in a single-patch device, which should allow for the continuous monitoring of the standard 12-lead ECG without requiring limb contact, both in daily life and in clinical practice.

Unconstrained Sleep Stage Estimation Based on Respiratory Dynamics and Body Movement.

OBJECTIVES: The aim of this study is to establish a sleep monitoring method that can classify sleep into four stages in an unconstrained manner using a polyvinylidene fluoride (PVDF) sensor for continuous and accurate estimation of sleep stages. METHODS: The study participants consisted of 12 normal subjects and 13 obstructive sleep apnea (OSA) patients. The physiological signals of the subjects were unconstrainedly measured using the PVDF sensor during polysomnography. The respiration and body movement signals were extracted from the PVDF data. Rapid eye movement (REM) sleep was estimated based on the average rate and variability of the respiratory signal. Wakefulness was detected based on the body movement signal. Variability of the respiratory rate was chosen as an indicator for slow-wave sleep (SWS) detection. Sleep was divided into four stages (wake, light, SWS, and REM) based on the detection results. RESULTS: The performance of the method was assessed by comparing the results with a manual scoring by a sleep physician. In an epoch-by-epoch analysis, the method classified the sleep stages with an average accuracy of 70.9 % and kappa statistics of 0.48. No significant differences were observed in the detection performance between the normal and OSA groups. CONCLUSIONS: The developed system and methods can be applied to a home sleep monitoring system.

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.

Defining a second epitope for heparin-induced thrombocytopenia/thrombosis antibodies using KKO, a murine HIT-like monoclonal antibody.

Heparin-induced thrombocytopenia/thrombosis (HIT/T) is a common complication of heparin therapy that is caused by antibodies to platelet factor 4 (PF4) complexed with heparin. The immune response is polyclonal and polyspecific, ie, more than one neoepitope on PF4 is recognized by HIT/T antibodies. One such epitope has been previously identified; it involves the domain between the third and fourth cysteine residues in PF4 (site 1). However, the binding sites for other HIT/T antibodies remain to be defined. To explore this issue, the binding site of KKO, an HIT/T-like murine monoclonal antibody, was defined. KKO shares a binding site with many HIT/T antibodies on PF4/heparin, but does not bind to site 1 or recognize mouse PF4/heparin. Therefore, the binding of KKO to a series of mouse/human PF4 chimeras complexed with heparin was examined. KKO recognizes a site that requires both the N terminus of PF4 and Pro34, which immediately precedes the third cysteine. Both regions lie on the surface of the PF4 tetramer in sufficient proximity (within 0.74 nm) to form a contiguous antigenic determinant. The 10 of 14 HIT/T sera that require the N terminus of PF4 for antigen recognition also require Pro34 to bind. This epitope, termed site 2, lies adjacent to site 1 in the crystal structure of the PF4 tetramer. Yet sites 1 and 2 can be recognized by distinct populations of antibodies. These studies further help to define a portion of the PF4 tetramer to which self-reactive antibodies develop in patients exposed to heparin.