Search

In this paper, we discuss accelerometry: an approach to human activity recognition and functional state assessment based on the usage of inertial sensors. Our discussion summarizes the pivotal aspects of accelerometric research, intending to provide an overview for the beginner practitioners in the field. Covered topics include signal processing techniques and machine learning algorithms, as well as general pointers to relevant literature.

Individualization of education at the university is key factor of digital transformation. Individualization of training involves considering the personal characteristics of students, providing them with services for choosing educational activities in the educational space of the university. There are technologies for construction of the educational space in now days: intelligent tutors; educational games; collaborative and project learning; personalized and adaptive learning; tools for learning analytics and educational data mining; ontologies for learning skills mining. Timetabling is key services for planning resources such as rooms, teachers, students. Multi-agent approach allows describe independent behavior of actors - student, dispatcher, course owner in self-organizing system of courses planning and selection. The model and services for events markup by the dispatcher and student courses selection and conflict resolving by the student are under review in this paper. The results of testing services at Ural Federal University are presented. The Ural Federal University has more than 36,000 students studying in 480 educational programs. The departments of 13 schools are involved in timetabling process four times in the year. The smart scheduling IT service is used for dispatch loads between 3000 teachers and thousands of rooms.

Scheduling is a well established process at universities. Each university has its own traditions and schedule quality parameters. The research about automated timetable are most common. Those researches uses a heuristic and meta heuristic approaches. There are many modifications of methods such local search methods, simulated annealing, adaptive memory programming, tabu searches and so on. But developing and implementation of IT services for Scheduling in university based on automated timetabling are well known problem. Last years the digital transformation of education is widely discussed topic. Key challenges are the radical individualization of education and lifelong education. And new models of educational process organization and resource use are under researcher’s focus. Adaptive learning, traditional, MOOC (Massive Open Online Courses) and blended courses, interschool and university shared courses are under point of view due those potential for individualization of education and lifelong education. The timetable model and real-world IT services are the basis for implementation such digital transformation in direction to increasing personalization and student motivation. The paper discusses the multi-agent scheduling model and the results of its implementation as services for interactive mode timetabling.

In this paper, we review the state of biofeedback methodology 50 years after its inception, with the objective to facilitate well-informed, scientific discourses around the topic. The scope of our review is limited to quantitative analysis of body of work published over the last decade, from which it is evident that biofeedback experiences a major methodological shift - away from strictly confined medical and research facilities, and towards personal, consumer-level use, thanks to wireless gadgets, wearables and smart devices available en masse. With such change come new opportunities and challenges, in the form of technological aspects, such as data processing trade-offs, efficiency of sensor placement, modalities and architectures, as well as general effectiveness of biofeedback in novel application areas, surrounded by privacy concerns and implications of e-health mobile technologies, which already available, but not yet ready for the prime-time. It is our intent to reflect on this liminal state of affairs.

Student trajectory - the main stages of training related to the choice: from admission to deduction. Student expulsion can be caused by various reasons. The main reasons are expulsion due to graduation from the university and academic failure. There are works that use various data to predict student expulsion: Demographic, Academic, Social and Behavioural in social networks. The peculiarity of this study is the use of data on the choice of a student in the formation of his individual trajectory. In paper the data model of the student and use-cases of the information system for students track prediction were considered. Impersonal data of 5000 bachelors from Ural Federal University formed a features space, including data on education, data on student track formation behavior and data on the social context. A model for predicting the trajectory of a student was developed. The proposed system can be adapted to the processing of data to form a competence portrait of student. On basic level, the system can be used to predict probability of a student being expelled from a University

This paper presents the modeling of an information system for collecting and processing data for the implementation and support of modular educational programs in the University. Features using modern educational technologies, including a personalized approach are described. The proposed system is called the “Digital Tutor” System. Use cases, functionality and service integration scheme are shown. The paper shows that this will allow informing students and receiving feedback from students about the learning conditions in the educational program.

This paper discusses the option of hardware-software integration, according to the assessment of human cognitive potential. The presented complex is based on an assessment of the integrative activity of the brain that allows one to get an idea of the current state of relations between the different areas of the brain structures (i.e., allows numerically estimate the number of structures involved in the activity). Evaluation of the integrative activity performed by a coherent analysis of the patient's EEG. The analysis is performed in real time using module coherent analysis developed in OpenVibe software.

In this research, we hypothesize that head-mounted accelerometric sensors can be effectively utilized in the tasks of motion activity classification and balance control evaluation, on a par with conventional methods based on stabilometric platforms. Following this hypothesis, we carried out a series of short experiments with stabilometric system "Stabilan-01-2" and consumer-grade tri-axial accelerometer. Collected motion signals were batch-processed and passed through feature extraction pipeline. After that, characteristic spaces were formed and classified using statistical machine learning methods. Results of the classification indicated that feature space of accelerometric data is informative enough to accurately classify motion activity associated with balance control, thus confirming our initial hypothesis. These results demonstrate that accelerometers can be used as a low-costly and portable alternative to stabilometric systems, and suggest a promising and novel approach to balance control assessment. © 2019 Author(s).

The Brain-Computer Interface (BCI) can be used for evaluation of the state of individuals during everyday routines. As shown in previous works, there is a relationship between the BCI multimodal metric with functional states of human. We have used power of Theta, Alpha, Beta low and Beta high electroencephalography rhythms and head motion data signals for multimodal metric. Heart Rate Variability (HRV) is common medical method for functional state assessment. In this paper the results of interrelation estimation between multimodal metric and HRV are shown. We used Pearson correlation coefficient (PCC) for estimates of interrelation between multimodal metric and HRV. It was found, the best results for estimates of parasympathetic part of the autonomic nervous system and suprasegmental regulation HRV have value of PCC more then critical value for Pearson correlation. © 2019 by SCITEPRESS - Science and Technology Publications, Lda. All rights reserved. The work was supported by Act 211 Government of the Russian Federation, contract № 02.A03.21.0006.

This paper discusses the possibility of development and testing a special methodology for the formation of a dynamic competence space. The main object of study is a student. The proposed methodology can take into account the risks of an erroneous assessment of the competences of an object, and it is also quickly adaptable and scalable. This paper presents a synthesis of methodology. Seven hypothetical ways of assessing and constructing skills space are considered.

In paper different standards for intelligent medical systems and multimodal architecture interfaces are under review. Application of Fast Healthcare Interoperability Resource (FHIR) standard for data model of intelligent multimodal interface are proposed. Intelligent multimodal interface for EEG and accelerometer bio-signals based on FHIR are shown. Use case diagrams, resource diagrams and class diagrams for production and research mode of this interface are described.

The estimation in real time of the functional and mental state level for the athlete during the loads is essential for management of the training process. New multimodal metric, obtained by means of the brain-computer interface (BCI), is proposed. The paper discusses the results of the joint usage of data from Emotiv EPOC+ mobile wireless headset. It includes motion sensors (accelerometer) and EEG channels. The features of the Emotiv EPOC+ interface allow to record the deviation of the head from the body axis, which provides an additional channel of information about the physical and mental (psycho-emotional) state of the athlete. Based on this data a new multimodal metric is calculated. Approbation of the metric was performed for functional stress studies on group of 10 volunteer subjects, including evaluations of the TOVA-test and the hyperventilation load. The joint application of different signals modalities allows to obtain estimates level of attention for these functional studies.

In the paper possibilities of functional research complex for mental status evaluation are shown. The complex consists of neuroheadset Emotiv Epoc+, including 14 channel wireless EEG-system with accelerometer, and ECG-recorder for acquisition of the heart rate variability signals (HRV). The record of biomedical signals was obtained in several functional states for 10 healthy subjects: Background (functional rest), TOVA-test (mental load), Hyperventilation (physical load) and Aftereffect (after test state). It was found that the power of the very low-frequency component of the HRV spectrum was significantly changed, which is caused by infiuence of the suprasegmentally regulation level on the heart rhythm. Also, the decrease in the control effect of the cerebral cortex (α-rhythm activity) is registered. At the same time, the amplitude of the average accelerometer values increases as result of uncontrolled movements in order to find the most stable position of the head relative to the gravitational vector.

Results of information analysis of radio brightness temperature fluctuations within the frequency range 650-850 MHz in brain tissues are described. For periods 10-20 and 60-70 sec, this method differentiates between apparently healthy subjects and patients with arterial hypertension or vegetative-vascular dystonia. © 2012 Springer Science+Business Media, Inc.