Noncontact tremor characterization using low-power wideband radar technology

Markerless Radio Frequency Indoor Monitoring for Telemedicine: Gait Analysis, Indoor Positioning, Fall Detection, Tremor Analysis, Vital Signs and Sleep Monitoring

Quantitative indoor monitoring, in a low-invasive and accurate way, is still an unmet need in clinical practice. Indoor environments are more challenging than outdoor environments, and are where patients experience difficulty in performing activities of daily living (ADLs). In line with the recent trends of telemedicine, there is an ongoing positive impulse in moving medical assistance and management from hospitals to home settings. Different technologies have been proposed for indoor monitoring over the past decades, with different degrees of invasiveness, complexity, and capabilities in full-body monitoring. The major classes of devices proposed are inertial-based sensors (IMU), vision-based devices, and geomagnetic and radiofrequency (RF) based sensors. In recent years, among all available technologies, there has been an increasing interest in using RF-based technology because it can provide a more accurate and reliable method of tracking patients' movements compared to other methods, such as camera-based systems or wearable sensors. Indeed, RF technology compared to the other two techniques has higher compliance, low energy consumption, does not need to be worn, is less susceptible to noise, is not affected by lighting or other physical obstacles, has a high temporal resolution without a limited angle of view, and fewer privacy issues. The aim of the present narrative review was to describe the potential applications of RF-based indoor monitoring techniques and highlight their differences compared to other monitoring technologies.

Hash Transformation and Machine Learning-Based Decision-Making Classifier Improved the Accuracy Rate of Automated Parkinson's Disease Screening

Digitalized hand-drawn pattern is a noninvasive and reproducible assistive manner to obtain hand actions and motions for evaluating functional tremors and upper-limb movement disorders. In this study, spirals and straight lines in polar coordinates are used to extract polar expression features such as the key parameters deviation (cm) and accumulation angle (rad). These parameters are quantitative manner to scale the variations of functional tremors in normal control subjects and patients with Parkinson's disease (PD) and essential tremor (ET). However, difficulty arises in using nonlinear polar expression features in the two-dimensional feature space to separate normal control subjects from those with PD and ET. To solve the nonlinear separable classification problem, hash transformation is used to map polar expression features to a high-dimensional space using hash weighing function and modulo operation. Then, a machine learning method, such as the generalized regression neural network (GRNN), is implemented to train a decision-making classifier using the particle swarm optimization (PSO) algorithm for possible class assessment. With the enrolled data from 50 subjects, the fivefold cross validation, mean true positive, mean true negative, and mean hit rates of 98.93%, 98.96%, and 98.93%, respectively, are obtained to quantify the performance of the proposed decision-making classifier to identify normal controls and subjects with PD or ET. The experimental results indicate that the proposed screening model can improve the accuracy rate compared with the conventional machine learning classifier.

Radar based technology for non-contact monitoring of accumulation of blood in the head: A numerical study

BACKGROUND

This theoretical study examines the use of radar to continuously monitor "accumulation of blood in the head" (ACBH) non-invasively and from a distance, after the location of a hematoma or hemorrhage in the brain was initially identified with conventional medical imaging. Current clinical practice is to monitor ABCH with multiple, subsequent, conventional medical imaging. The radar technology introduced in this study could provide a lower cost and safe alternative to multiple conventional medical imaging monitoring for ACBH.

MATERIALS AND METHODS

The goal of this study is to evaluate the feasibility of using radar to monitor changes in blood volume in the brain through a numerical simulation of ACBH monitoring from remote, with a directional spiral slot antennae, in 3-D models of the brain. The focus of this study is on evaluating the effect of frequencies on the antennae reading. To that end we performed the calculations for frequencies of 100 MHz, 500 MHz and 1 GHz.

RESULTS AND DISCUSSION

The analysis shows that the ACBH can be monitored with radar and the monitoring resolution improves with an increase in frequency, in the range studied. However, it also appears that when typical clinical dimensions of hematoma and hemorrhage are used, the variable ratio of blood volume radius and radar wavelength can bring the measurements into the Mie and Rayleigh regions of the radar cross section. In these regions there is an oscillatory change in signal with blood volume size. For some frequencies there is an increase in signal with an increase in volume while in others there is a decrease.

CONCLUSIONS

While radar can be used to monitor ACBH non-invasively and from a distance, the observed Mie region dependent oscillatory relation between blood volume size and wavelength requires further investigation. Classifiers, multifrequency algorithms or ultra-wide band radar are possible solutions that should be explored in the future.

A Real-Time Kinect Signature-Based Patient Home Monitoring System

Assessment of body kinematics during performance of daily life activities at home plays a significant role in medical condition monitoring of elderly people and patients with neurological disorders. The affordable and non-wearable Microsoft Kinect ("Kinect") system has been recently used to estimate human subject kinematic features. However, the Kinect suffers from a limited range and angular coverage, distortion in skeleton joints' estimations, and erroneous multiplexing of different subjects' estimations to one. This study addresses these limitations by incorporating a set of features that create a unique "Kinect Signature". The Kinect Signature enables identification of different subjects in the scene, automatically assign the kinematics feature estimations only to the subject of interest, and provide information about the quality of the Kinect-based estimations. The methods were verified by a set of experiments, which utilize real-time scenarios commonly used to assess motor functions in elderly subjects and in subjects with neurological disorders. The experiment results indicate that the skeleton based Kinect Signature features can be used to identify different subjects in high accuracy. We demonstrate how these capabilities can be used to assign the Kinect estimations to the Subject of Interest, and exclude low quality tracking features. The results of this work can help in establishing reliable kinematic features, which can assist in future to obtain objective scores for medical analysis of patient condition at home while not restricted to perform daily life activities.

Human behavioral assessments in current research of Parkinson's disease

Parkinson's disease (PD) is traditionally classified as a movement disorder because patients mainly complain about motor symptoms. Recently, non-motor symptoms of PD have been recognized by clinicians and scientists as early signs of PD, and they are detrimental factors in the quality of life in advanced PD patients. It is crucial to comprehensively understand the essence of behavioral assessments, from the simplest measurement of certain symptoms to complex neuropsychological tasks. We have recently reviewed behavioral assessments in PD research with animal models (Asakawa et al., 2016). As a companion volume, this article will systematically review the behavioral assessments of motor and non-motor PD symptoms of human patients in current research. The major aims of this article are: (1) promoting a comparative understanding of various behavioral assessments in terms of the principle and measuring indexes; (2) addressing the major strengths and weaknesses of these behavioral assessments for a better selection of tasks/tests in order to avoid biased conclusions due to inappropriate assessments; and (3) presenting new concepts regarding the development of wearable devices and mobile internet in future assessments. In conclusion we emphasize the importance of improving the assessments for non-motor symptoms because of their complex and unique mechanisms in human PD brains.

Human body parts tracking and kinematic features assessment based on RSSI and inertial sensor measurements

Acquisition of patient kinematics in different environments plays an important role in the detection of risk situations such as fall detection in elderly patients, in rehabilitation of patients with injuries, and in the design of treatment plans for patients with neurological diseases. Received Signal Strength Indicator (RSSI) measurements in a Body Area Network (BAN), capture the signal power on a radio link. The main aim of this paper is to demonstrate the potential of utilizing RSSI measurements in assessment of human kinematic features, and to give methods to determine these features. RSSI measurements can be used for tracking different body parts' displacements on scales of a few centimeters, for classifying motion and gait patterns instead of inertial sensors, and to serve as an additional reference to other sensors, in particular inertial sensors. Criteria and analytical methods for body part tracking, kinematic motion feature extraction, and a Kalman filter model for aggregation of RSSI and inertial sensor were derived. The methods were verified by a set of experiments performed in an indoor environment. In the future, the use of RSSI measurements can help in continuous assessment of various kinematic features of patients during their daily life activities and enhance medical diagnosis accuracy with lower costs.