Fall Prediction in People with Parkinson's Disease.

A preliminary study result predicting fall events in patients with Parkinson's disease (PD) by using a simple motion sensor is described in this paper. Causes of falls in people with PD can be postural instability, freezing of gait, festinating gait, dyskinesias, visuospatial dysfunction, orthostatic hypotension, and posture problems. This study uses only one motion sensor in collecting data. Thus, only fall events caused by festinating gait factors, which are moments when the patient suddenly moves faster with smaller steps, can be performed and tested. In this preliminary study, fall event scenarios of simulated test cases are performed by five healthy young subjects aged 20 to 28 years old. The acceleration mode in the motion sensor provides information that can detect how fast the subjects move. Data collected by the sensor will be analyzed by simple analysis methods and machine learning techniques classification. The proposed study achieved an accuracy of 70.3% for the 10-class model, while for binary classification, the accuracy was 99%. Clinical Relevance-This study focuses on predicting falls by analyzing the gaits prior to an actual so that fall prediction can be possible. If falls can be predicted, researchers can develop other protective gear to prevent fall-related injuries not only for PD patients but also for the elderly.

Classification of Handwashing and Similar Activities.

In this paper, an ensemble gentle boost decision tree classification algorithm is trained to classify handwashing from similar activities such as applying lotion to hands. Data is collected using a 3-axis accelerometer and gyroscope worn on the wrist. First, the data collection procedure is described. Then, feature identification is discussed. Once the feature matrix was created, the MATLAB classification learner app was used to classify the data based on the identified features. The overall classification rate achieved was 91.6% using an optimized boosted ensemble classifier. Clinical Relevance- The spreading of germs could be prevented by simple activities such as proper handwashing and wearing masks during the pandemic. This research shows that wearable sensors with machine learning algorithms can alert the users and guide users to wash their hands properly.

Deciphering clock cell network morphology within the biological master clock, suprachiasmatic nucleus: From the perspective of circadian wave dynamics.

The biological master clock, suprachiasmatic nucleus (of rat and mouse), is composed of ~10,000 clock cells which are heterogeneous with respect to their circadian periods. Despite this inhomogeneity, an intact SCN maintains a very good degree of circadian phase (time) coherence which is vital for sustaining various circadian rhythmic activities, and it is supposedly achieved by not just one but a few different cell-to-cell coupling mechanisms, among which action potential (AP)-mediated connectivity is known to be essential. But, due to technical difficulties and limitations in experiments, so far very little information is available about the morphology of the connectivity at a cellular scale. Building upon this limited amount of information, here we exhaustively and systematically explore a large pool (~25,000) of various network morphologies to come up with some plausible network features of SCN networks. All candidates under consideration reflect an experimentally obtained 'indegree distribution' as well as a 'physical range distribution of afferent clock cells.' Then, importantly, with a set of multitude criteria based on the properties of SCN circadian phase waves in extrinsically perturbed as well as in their natural states, we select out appropriate model networks: Some important measures are, 1) level of phase dispersal and direction of wave propagation, 2) phase-resetting ability of the model networks subject to external circadian forcing, and 3) decay rate of perturbation induced "phase-singularities." The successful, realistic networks have several common features: 1) "indegree" and "outdegree" should have a positive correlation; 2) the cells in the SCN ventrolateral region (core) have a much larger total degree than that of the dorsal medial region (shell); 3) The number of intra-core edges is about 7.5 times that of intra-shell edges; and 4) the distance probability density function for the afferent connections fits well to a beta function. We believe that these newly identified network features would be a useful guide for future explorations on the very much unknown AP-mediated clock cell connectome within the SCN.

Garment Integrated Spinal Posture Detection Using Wearable Magnetic Sensors.

Spine Curvature Disorder (SCD) is a medical condition that affects the shape of the spine. Methods of monitoring SCDs involve visual inspection followed by X-rays and measurements. Once a patient is diagnosed with SCD and treatment or therapy is implemented, progress is tracked by exposing the patient to multiple periodic X-rays to determine the spine responses to treatments or therapies. Multiple exposures to X-rays is not desirable and is also costly. Therefore, we propose a new method for detecting and monitoring SCD and present our initial research results. We are implementing a non-invasive method that can detect and monitor the spinal postures of SCD. Magnets are placed on a shirt a grid form then a sensor system would be placed on the chest of the body. An on-body magnetic sensor records the sensor data values to determine if the upper body posture is straight or is curved which in turn can assist in detecting if the spine is deformed. We present our initial results on magnetic sensor testing and preliminary results using wearable sensors and a garment integrated magnetic shirt.

Diversity in the structure of action potential-mediated neural connectivity within rat supra chiasmatic nucleus.

Action potential (AP)-mediated cell-to-cell communication is essential for the frequency-locking and phase-synchronization of the clock cells within the biological master clock, suprachiasmatic nucleus (SCN). Nevertheless, the morphology of its network connectivity is largely unexplored. Here, with an optimized optogenetic light-stimulation and scanning protocol, we report some key characteristics of the inhibitory receptive field (IRF), the area which brings inhibitory synaptic currents to a given target cell, and basic statistics of the inhibitory network connections of rat SCN clock cells. ChR2 transfected, slice cultures of rat SCN were stimulated by a blue power LED light in a repetitive box-scanning modes, while a target cell was whole-cell patched. The registered inhibitory postsynaptic currents, which were brought by light-induced APs of presynaptic neurons, were mostly GABAergic. The sizes and shapes of IRFs of SCN cells were very diverse, and the number of presynaptic cells making up the IRF of a given target cell followed an exponential distribution with an average value of 8.9 approximately, according to our clustering analysis which is based on a hybrid measure D, combining the physical distance r and the difference in the current amplitudes of two different sites. Although this estimate inevitably depends on the construct of the measure D, it is found not so sensitive on the parameter w, which weighs the relative significance of the current amplitude different with respect to the physical distance r: The average number of presynaptic neurons varies < 26% over a significant range of 0 < w < 30. On average, the presynaptic connection number density around a target cell falls off as an exponentially decreasing function of r. But, its space constant (~210.7 µm) is quite large that long-range (>210.7 µm) neural connections are abundant (>66.9%) within the SCN.

Training a Neural Network for Vocal Stereotypy Detection.

This system was developed to detect and diagnose vocal stereotypies made by non-verbal autistic children. Vocal stereotypies are loud non-speech vocalizations made by these children.The system discussed in this paper uses a deep learning neural network to detect these vocalizations. Using similar data from other recorded human voices, the system can be trained to detect the non-speech vocalizations of autistic children. By detecting vocalizations the proposed system can be used to recognize the stimming of non-verbal autistic children from the background noise.

Vocal Stereotypy Detection: An Initial Step to Understanding Emotions of Children with Autism Spectrum Disorder.

A system has been developed to automatically record and detect behavioral patterns and vocal stereotypy which is also known as vocal stimming, a non-verbal vocalization often observed in children with Autism Spectrum Disorder (ASD). The system incorporates audio, video and wearable accelerometer based sensors. Microphones and video camera were used to collect data and were used for analysis. KSVD, which is a generalized version of the k-means clustering algorithms for dictionary learning, was used to detect vocal stereotypy. Observing the subspace that the data lives in allows us to detect vocal stimming and sounds of frustration. The proposed system was able to detect vocalized stimming with detection rate between 73 - 93 percent.

Semi-supervised event detection using higher order statistics for multidimensional time series accelerometer data.

In this study, we target to automatically detect stereotypical behavioral patterns (stereotypy) and self-injurious behaviors (SIB) of Autistic children which can lead to critical damages or wounds as they tend to repeatedly harm oneself. Our custom designed accelerometer based wearable sensors are placed at wrists, ankles and upper body to detect stereotypy and SIB. The analysis was done on four children diagnosed with ASD who showed repeated behaviors that involve part of the body such as flapping arms, body rocking and self-injurious behaviors such as punching their face, or hitting their legs. Our goal of detecting novel events relies on the fact that the limitation of training data and variability in the possible combination of signals and events also make it impossible to design a single algorithm to understand all events in natural setting. Therefore, a semi-supervised method to discover and track unknown events in a multidimensional sensor data rises as a very important topic in classification and detection problems. In this paper, we show how the Higher Order Statistics (HOS) features can be used to design dictionaries and to detect novel events in a multichannel time series data. We explain our methods to detect novel events in a multidimensional time series data and combine the proposed semi-supervised learning method to improve the adaptability of the system while maintaining comparable detection accuracy as the supervised method. We, compare our results to the supervised methods that we have previously developed and show that although semi-supervised method do not achieve better performance compared to supervised methods, it can efficiently find new events and anomalies in multidimensional time series data with similar performance of the supervised method. We show that our proposed method achieves recall rate of 93.3% compared to 94.1% for the supervised method studied earlier.

Automatic characterization and detection of behavioral patterns using linear predictive coding of accelerometer sensor data.

In this study, we target to automatically detect behavioral patterns of patients with autism. Many stereotypical behavioral patterns may hinder their learning ability as a child and patterns such as self-injurious behaviors (SIB) can lead to critical damages or wounds as they tend to repeatedly harm one single location. Our custom designed accelerometer based wearable sensor can be placed at various locations of the body to detect stereotypical self-stimulatory behaviors (stereotypy) and self-injurious behaviors of patients with Autism Spectrum Disorder (ASD). A microphone was used to record sounds so that we may understand the surrounding environment and video provided ground truth for analysis. The analysis was done on four children diagnosed with ASD who showed repeated self-stimulatory behaviors that involve part of the body such as flapping arms, body rocking and self-injurious behaviors such as punching their face, or hitting their legs. The goal of this study is to devise novel algorithms to detect these events and open possibility for design of intervention methods. In this paper, we have shown time domain pattern matching with linear predictive coding (LPC) of data to design detection and classification of these ASD behavioral events. We observe clusters of pole locations from LPC roots to select candidates and apply pattern matching for classification. We also show novel event detection using online dictionary update method. We show that our proposed method achieves recall rate of 95.5% for SIB, 93.5% for flapping, and 95.5% for rocking which is an increase of approximately 5% compared to flapping events detected by using wrist worn sensors in our previous study.

Optimal sensor location for body sensor network to detect self-stimulatory behaviors of children with Autism Spectrum Disorder.

In this study, we investigate various locations of sensor positions to detect stereotypical self-stimulatory behavioral patterns of children with Autism Spectrum Disorder (ASD). The study is focused on finding optimal detection performance based on sensor location and number of sensors. To perform this study, we developed a wearable sensor system that uses a 3 axis accelerometer. A microphone was used to understand the surrounding environment and video provided ground truth for analysis. The recordings were done on 2 children diagnosed with ASD who showed repeated self-stimulatory behaviors that involve part of the body such as flapping arms, body rocking and vocalization of non-word sounds. We used time-frequency methods to extract features and sparse signal representation methods to design over-complete dictionary for data analysis, detection and classification of these ASD behavioral events. We show that using single sensor on the back achieves 95.5% classification rate for rocking and 80.5% for flapping. In contrast, flapping events can be recognized with 86.5% accuracy using wrist worn sensors.

Classification of continuously executed early morning activities using wearable wireless sensors.

In this paper, we study the personal monitoring system that classifies the continuously executed early morning activities of daily living. The system is intended to assist those with cognitive impairments due to traumatic brain injuries. The system can be used to help therapists in hospitals or could be deployed in one's home to track and monitor the activities executed by the recovering patients. We begin by briefly describing the infrastructure of our cost-effective system which uses fixed and wearable wireless sensors and show results related to the detection of activities continuously executed in the morning. Both frequency and time domain features from an accelerometer attached to the right wrist were extracted and used for classification using Gaussian mixture models, followed by a finite state machine. We show promising classification results obtained from 5 subjects. Overall classification rate is 88.3 % for 4 activities of interests.