Search

This paper concerns the design, analysis, and simulation of a 3D non-stationary channel model fed with inertial measurement unit (IMU) data. The work in this paper provides a framework for simulating the micro-Doppler signatures of indoor channels for human activity recognition by using radio-frequency-based sensing technologies. The major human body segments, such as wrists, ankles, torso, and head, are modelled as a cluster of moving point scatterers. We provide expressions for the time variant (TV) speed and TV angles of motion based on 3D trajectories of the moving person. Moreover, we present mathematical expressions for the TV Doppler shifts and TV path gains associated with each moving point scatterer. Furthermore, a model of the non-stationary time variant channel transfer function (TV-CTF) is provided, which takes into account the effects caused by a moving person as well as fixed objects, such as furniture, walls, and ceiling. The micro-Doppler signatures of the moving person is extracted from the TV-CTF by employing the concept of the spectrogram, whose expression is also provided in closed form. Our model is confirmed by channel state information (CSI) measurements taken during walking, falling, and sitting activities. The proposed channel model is fed with IMU data that has been collected. We evaluate the micro-Doppler signature of the model and CSI measurements. The results show a good agreement between the spectrograms and the TV mean Doppler shifts of our IMU-driven channel model and the measured CSI. The proposed model enables a paradigm shift from traditional experimental-based approaches to future simulation-based approaches for the design of human activity recognition systems.

Robust and accurate human activity recognition (HAR) systems are essential to many human-centric services within active assisted living and healthcare facilities. Traditional HAR systems mostly leverage a single sensing modality (e.g., either wearable, vision, or radio frequency sensing) combined with machine learning techniques to recognize human activities. Such unimodal HAR systems do not cope well with real-time changes in the environment. To overcome this limitation, new HAR systems that incorporate multiple sensing modalities are needed. Multiple diverse sensors can provide more accurate and complete information resulting in better recognition of the performed activities. This article presents WiWeHAR-a multimodal HAR system that uses combined Wi-Fi and wearable sensing modalities to simultaneously sense the performed activities. WiWeHAR makes use of standard Wi-Fi network interface cards to collect the channel state information (CSI) and a wearable inertial measurement unit (IMU) consisting of accelerometer, gyroscope, magnetometer sensors to collect the user's local body movements. We compute the time-variant mean Doppler shift (MDS) from the processed CSI data and magnitude from the inertial data for each sensor of the IMU. Thereafter, we separately extract various time- and frequency-domain features from the magnitude data and the MDS. We apply feature-level fusion to combine the extracted features, and finally supervised learning techniques are used to recognize the performed activities. We evaluate the performance of WiWeHAR by using a multimodal human activity data set, which was obtained from 9 participants. Each participant carried out four activities, such as walking, falling, sitting, and picking up an object from the floor. Our results indicate that the proposed multimodal WiWeHAR system outperforms the unimodal CSI, accelerometer, gyroscope, and magnetometer HAR systems and achieves an overall recognition accuracy of 99.6%-100%.

Unconscious mental processes have recently started gaining attention in a number of scientific disciplines. One of the theoretical frameworks for describing unconscious processes was introduced by Jung as a part of his model of the psyche. This framework uses the concept of archetypes that represent prototypical experiences associated with objects, people, and situations. Although the validity of Jungian model remains an open question, this framework is convenient from the practical point of view. Moreover, archetypes found numerous applications in the areas of psychology and marketing. Therefore, observation of both conscious and unconscious traces related to archetypal experiences seems to be an interesting research endeavor. In a study with 36 subjects, we examined the effects of experiencing conglomerations of unconscious emotions associated with various archetypes on the participants' introspective reports and patterns of physiological activations. Our hypothesis for this experiment was that physiological data may predict archetypes more precisely than introspective reports due to the implicit nature of archetypal experiences. Introspective reports were collected using the Self-Assessment Manikin (SAM) technique. Physiological measures included cardiovascular, electrodermal, respiratory responses and skin temperature of the subjects. The subjects were stimulated to feel four archetypal experiences and four explicit emotions by means of film clips. The data related to the explicit emotions served as a reference in analysis of archetypal experiences. Our findings indicated that while prediction models trained on the collected physiological data could recognize the archetypal experiences with accuracy of 55 percent, similar models built based on the SAM data demonstrated performance of only 33 percent. Statistical tests enabled us to confirm that physiological observations are better suited for observation of implicit psychological constructs like archetypes than introspective reports.

AbstractIn assisted ambulation, the user’s psychological comfort has a significant impact not only on acceptability of mobility aids but also on overall gait performance. Specifically, in the case of rollators, negative states such as distress may result in balance loss, inefficient manoeuvres, and an increased risk of falling. This paper presents a pilot study to investigate the effect of distress on rollator assisted navigation. To achieve this goal, a novel test protocol is proposed to assess distress while walking with a rollator, using the Self-Assessment Manikin (SAM) questionnaire. First, the participant completes a standardised visual stress induction test and fills in a SAM questionnaire on the dimensions of arousal and valence, to establish personal benchmarks. Then, they complete a course consisting of four navigation tasks with different levels of difficulty that affect the rollator manoeuvrability, filling in a SAM questionnaire after each task. An experiment including 25 healthy volunteers has been completed. Our preliminary results show that stressors like uneven or sloping surfaces increase perceived stress, whereas the shape of the trajectory does not significantly affect stress. The ultimate purpose of this work is to validate a performance-oriented protocol to investigate the dynamics of stress response in assisted walk and to train automatic stress detection systems.

In the past decade, the use of wearable medical devices has been a great breakthrough in clinical practice, trials, and research. In the Parkinson's disease field, clinical evaluation is time limited, and healthcare professionals need to rely on retrospective data collected through patients' self-filled diaries and administered questionnaires. As this often leads to inaccurate evaluations, a more objective system for symptom monitoring in a patient's daily life is claimed. In this regard, the use of wearable medical devices is crucial. This study aims at presenting a review on STAT-ONTM, a wearable medical device Class IIa, which provides objective information on the distribution and severity of PD motor symptoms in home environments. The sensor analyzes inertial signals, with a set of validated machine learning algorithms running in real time. The device was developed for 12 years, and this review aims at gathering all the results achieved within this time frame. First, a compendium of the complete journey of STAT-ONTM since 2009 is presented, encompassing different studies and developments in funded European and Spanish national projects. Subsequently, the methodology of database construction and machine learning algorithms design and development is described. Finally, clinical validation and external studies of STAT-ONTM are presented.

Among Parkinson's disease (PD) symptoms, freezing of gait (FoG) is one of the most debilitating. To assess FoG, current clinical practice mostly employs repeated evaluations over weeks and months based on questionnaires, which may not accurately map the severity of this symptom. The use of a non-invasive system to monitor the activities of daily living (ADL) and the PD symptoms experienced by patients throughout the day could provide a more accurate and objective evaluation of FoG in order to better understand the evolution of the disease and allow for a more informed decision-making process in making adjustments to the patient's treatment plan. This paper presents a new algorithm to detect FoG with a machine learning approach based on Support Vector Machines (SVM) and a single tri-axial accelerometer worn at the waist. The method is evaluated through the acceleration signals in an outpatient setting gathered from 21 PD patients at their home and evaluated under two different conditions: first, a generic model is tested by using a leave-one-out approach and, second, a personalised model that also uses part of the dataset from each patient. Results show a significant improvement in the accuracy of the personalised model compared to the generic model, showing enhancement in the specificity and sensitivity geometric mean (GM) of 7.2%. Furthermore, the SVM approach adopted has been compared to the most comprehensive FoG detection method currently in use (referred to as MBFA in this paper). Results of our novel generic method provide an enhancement of 11.2% in the GM compared to the MBFA generic model and, in the case of the personalised model, a 10% of improvement with respect to the MBFA personalised model. Thus, our results show that a machine learning approach can be used to monitor FoG during the daily life of PD patients and, furthermore, personalised models for FoG detection can be used to improve monitoring accuracy.

Human movement analysis is a field of wide interest since it enables the assessment of a large variety of variables related to quality of life. Human movement can be accurately evaluated through Inertial Measurement Units (IMU), which are wearable and comfortable devices with long battery life. The IMU’s movement signals might be, on the one hand, stored in a digital support, in which an analysis is performed a posteriori. On the other hand, the signal analysis might take place in the same IMU at the same time as the signal acquisition through online classifiers. The new sensor system presented in this paper is designed for both collecting movement signals and analyzing them in real-time. This system is a flexible platform useful for collecting data via a triaxial accelerometer, a gyroscope and a magnetometer, with the possibility to incorporate other information sources in real-time. A µSD card can store all inertial data and a Bluetooth module is able to send information to other external devices and receive data from other sources. The system presented is being used in the real-time detection and analysis of Parkinson’s disease symptoms, in gait analysis, and in a fall detection system.

Altered movement control is typically the first noticeable symptom manifested by Parkinson’s disease (PD) patients. Once under treatment, the effect of the medication is very patent and patients often recover correct movement control over several hours. Nonetheless, as the disease advances, patients present motor complications. Obtaining precise information on the long-term evolution of these motor complications and their short-term fluctuations is crucial to provide optimal therapy to PD patients and to properly measure the outcome of clinical trials. This paper presents an algorithm based on the accelerometer signals provided by a waist sensor that has been validated in the automatic assessment of patient’s motor fluctuations (ON and OFF motor states) during their activities of daily living. A total of 15 patients have participated in the experiments in ambulatory conditions during 1 to 3 days. The state recognised by the algorithm and the motor state annotated by patients in standard diaries are contrasted. Results show that the average specificity and sensitivity are higher than 90%, while their values are higher than 80% of all patients, thereby showing that PD motor status is able to be monitored through a single sensor during daily life of patients in a precise and objective way.

In this paper, we present a novel tool to measure engagement in people with dementia playing board games and interacting with a social robot, Pleo. We carried out two studies to reach a comprehensive inventory of behaviours accounting for engagement in dementia. The first one is an exploratory study aimed at modelling engagement in cognitive board games. The second one is a longitudinal study to investigate how people with dementia express engagement in cognitive games and in interactions with social robots. We adopted a technique coming from Ethology to mould behaviour, the ethogram. Ethogram is founded on low level behaviours, and allows hierarchical structuring. Herein, we present preliminary results consisting in the description of two ethograms and in their structuring obtained through thematic analysis. Such results show that an underlying structure of engagement exists across activities, and that different activities trigger different behavioural displays of engagement that adhere to such a structure.