Fig 4 - uploaded by Vitalij Denisov
Content may be subject to copyright.
Interpretation of data samples of 7 channels c 3, c 4, p 3, p 4, o 1, o 2, and EEG taken at 250 Hz for 10 seconds, for 7x2500 samples from binary Matlab mat-file 

Interpretation of data samples of 7 channels c 3, c 4, p 3, p 4, o 1, o 2, and EEG taken at 250 Hz for 10 seconds, for 7x2500 samples from binary Matlab mat-file 

Source publication
Article
Full-text available
An approach of interpreting some bioinformatics data using Self-Organizing Map (SOM) type neural networks is described. The ways are proposed constructing intelligent program tools of embedded agents to collect ECG and EEG data for academic usage in a virtual e-laboratory. A SOM-based diagnostic algorithm is proposed interpreting data from Medical...

Context in source publication

Context 1
... of each microcontroller. Every Bio-Robot i=1,m in an e- laboratory is controlled by its own JADEX-based bio-robot control agent JBRCA i=1,m . Multi-robot control strategy by Multi-user is being permanently realized by adaptive multi-robot control agents, the MRCA.. Each Bio- Robot i=1,m is controlled directly by its IRC Client i=1,m via serial RS232 port using HyperTerminal software. Adaptive control strategy of Bio-Robot i=1,m can be implemented only having its online reprogramming capabilities by permanently using In System Programming (ISP) feature of each microcontroller. In the e-laboratory of Fig. 1, this is realized by arranging for each Bio-Robot i=1,m both the Commu- tative Hardware i=1,m and ISP Programmer i=1,m . These additional elements allow communication with each remote ATmega8-based embedded agent either for its control or reprogramming purposes implementing adaptive properties of Bio-Robot i=1,m . A learner – a multi-user type client is able to run e-laboratory experiments by an aid of the MRCA and IRC protocol. This is realized by using two type programming Delphi and Java based JADEX environments. This approach allows to realize software type communication between MRCA and Bio-Robot Control Agent JBRCA i=1,m via TCP/IP sockets. The JBRCA i=1,m also communicates with KDB – the dynamically changing knowledge base of an e-laboratory. We acknowledge the Computer Science Department at Colorado State University in our communications of their work “Classification of Electroencephalogram (EEG) Signals for Brain-Machine Interfaces” by [16] involving their EEG data recorded by Zak Keirn at Purdue University for his work on his Masters of Science thesis in the Electrical Engineering Department at Purdue. This data is available as a 23 MB, from binary Matlab mat-file [10, 18]. The data is a cell array of cell arrays. Each individual cell array is made up of a subject string, task string, trial string, and data array. Each data array is 7 rows by 2500 columns. The 7 rows correspond to channels c 3, c 4, p 3, p 4, o 1, o 2, and ECG of the 10-20 System (Fig. 2). There is a standardized EEG electrodes placement system called the 10-20 System [14, 15]. The 10-20 System is based on the relationship between the location of an electrode and the underlying area of cerebral cortex. Each point on it indicates a possible electrode position. Each site has a letter (to identify the lobe) and a number or another letter to identify the hemisphere location. The letters F, T, C, P, and O stand for Frontal, Temporal, Central, Parietal and Occipital. (Note that there is no “central lobe”, but this is just used for identification purposes.) Even numbers (2, 4, 6, and 8) refer to the right hemisphere and odd numbers (1, 3, 5, 7) refer to the left hemisphere. The z refers to an electrode placed on the midline. Also note that the smaller the number, the closer the position is to the midline. In our case, across columns are samples taken at 250 Hz for 10 seconds, for 2500 samples (Fig. 3–4). For example, the first cell array of Fig. 3 represents data of subject 1 who completed task 1 – The Baseline Task under the trial 1 for 10 seconds. This data may look like a plot in Fig. 4. Recordings were made by Zak Keirn at Purdue University with reference to electrically linked mastoids A1 and A2 of Fig. 2. EEG was recorded between the forehead above the left brow line and another on the left cheekbone. Recording was performed with a bank of Grass 7P511 amplifiers whose band pass analog filters were set at 0.1 to 100 Hz. Subjects 1 and 2 were employees of a university and were left-handed age 48 and right-handed age 39, respectively. Subjects 3 through 7 were right- handed college students between the age of 20 and 30 years old. All were mail subjects with the exception of Subject 5. Subjects performed five trials of each task in one day. They returned to do a second five trials on another day. Subjects 2 and 7 completed only one 5-trial ses- sion. Subject 5 completed three sessions. The mental tasks and their labels are described in Table 1. The recorded EEG Data from [15] was preprocessed by constructing separate arrays of different length vectors. The last attrib- ute of each vector in every array was a label A through E interpreting a mental task performed by a subject in accor- dance with the Table 1. The arrays were constructed for vectors of 15 th order representing a sampling interval of 0.06 seconds through vectors of 1000 th order with sampling interval of 4 seconds in recorded data. The SOM networks for 226 th order vectors are given in Fig. 5. The method of label prescription for the SOM clusters is applied constructing right part of Fig. 5. It allows interpreting the clusters by giving them the names A, B, C, D, and E of the mental tasks described in Table 1. For data interpretation during a process of constructing a courseware in an e- laboratory, the results of visual indication of some mental states can be used. This indication of subject’s mental states can be seen on the SOM’s of Fig. 8. The right part SOM of Fig. 8 shows that the clusters representing the mental task B are located at the corners of the map. We can also see from the left part SOM of Fig. 5 that the mental task B (Math Task) is represented by higher values of ...

Similar publications

Conference Paper
Full-text available
A novel approach to enhance the robustness of handovers in LTE femtocells is presented. A modified Self Organizing Map is used to allow femtocells to learn about their specific indoor environment including the locations that have prompted handover requests. Optimized handover parameter values are then used that are specific to these locations. This...
Article
Full-text available
In this paper we present a novel system for automatically marking up text documents into XML. The system uses the techniques of the Self-Organising Map (SOM) algorithm in conjunction with an inductive learning algorithm, C5.0. The SOM algorithm clusters the XML marked-up documents on a two-dimensional map such that documents having similar content...
Article
Full-text available
A novel approach to handover management for Long-Term Evolution (LTE) femtocells is presented. Within LTE, the use of self-organizing networks (SONs) is included as standard, and handover management is one of its use cases. Base stations can autonomously decide whether handover should take place and assign the values of relevant parameters. Due to...
Article
Full-text available
We introduce a novel two-stage automatic XML mark-up system, which combines the WEBSOM approach to document categorisation in conjunction with the C5 inductive learning algorithm. The WEBSOM method clusters the XML marked-up documents such that semantically similar documents lie close together on a Self-Organising Map (SOM). The C5 algorithm automa...
Conference Paper
Full-text available
In diesem Bericht wird eine robuste Methode zur direkten Klassifizierung texturierter 3D-Punkte komplexer urbaner Laserscanningszenen vorgestellt. Die Methode basiert auf den geometrischen und farblichen Charakteristika der Punkte, indem für jeden Punkt sein jeweiliges Fast Point Feature Histogramm, sein Zenitwinkel zur Flächennormalen sowie sein H...

Citations

... The control subsystem is working as on-line diagnostic systems of complex mechanisms with the cooperation of multi-agent's activities in recognizing human affect sensing. Being able both to provide an intelligent accident preventive robot-based support for people with moving disabilities and to include affect sensing in Human Computer Interaction (HCI, in providing e-health care for people with moving disabilities), Human-Robot Interaction (HRI, for assisting tele-healthcare patients remaining autonomous), and Computer Mediated Communication (CMC, in providing adaptive user-robot friendly collaboration), such system should depend upon the possibility of extracting emotions without interrupting the user during HCI, HRI, or CMC (Mandryk, Atkins 2007;Bielskis et al. 2007;Pentland 2004). Emotion is a mind-body phenomenon that is accessible at different levels of observation (social, psychological, cerebral and physiological). ...
Article
Full-text available
The aims of this research are focused on the construction of intellectualized equipments for people with moving disabilities to help them in sustainable integration into environment. The problem is to reveal main components of diagnosis of disabled persons, as well as to develop decision making models which are integrated into the control mechanisms of the special equipments, that are assigned to the class of bio‐robots. This paper analyses the approach of the construction of such type of bio‐robots with possibilities to integrate different knowledge representation techniques for the development of the reinforcement framework with multiple cooperative agents for the recognition of the diagnosis of emotional situation of disabled persons. Large‐scale of multidimensional recognitions of emotional diagnosis of disabled persons often generate a large amount of multi‐dimensional data with complex recognition mechanisms, based on the integration of different knowledge representation techniques and complex inference models. Sensors can easily record primary data; however, the recognition of abnormal situations, cauterisation of emotional stages and resolution for certain type of diagnosis is an oncoming issue for bio‐robot constructors. The research results present the development of multi‐layered model of this framework with the integration of the evaluation of fuzzy neural control of speed of two wheelchair type robots working in real time by providing moving support for disabled individuals. An approach for representation of reasoning processes, using fuzzy logical Petri nets for evaluation of physiological state of individuals is presented. The reasoning is based on recognition of emotions of persons during their activities. Santrauka Šio mokslinio tyrimo tikslai yra nukreipti į intektualizuotų įrenginių, skirtų žmonėms su judėjimo negalia ir užtikrinančių jų būklės stebėseną ir darnaus judėjimo valdymo aplinkoje galimybes, kūrimą. Sprendžiami uždaviniai skirti neįgaliųjų diagnozės pagrindinių komponenčių tyrimams, sudarant lanksčius sprendimų priėmimo modelius, integruojamus į specialių įrenginių valdymo mechanizmus, kurie priskiriami biorobotų klasei. Straipsnyje pateikiami metodai, kaip konstruoti tokio tipo biorobotų sistemas, leidžiant skirtingų žinių vaizdavimo priemones integruoti į sistemą, kad būtų sukurta daugelio agentų bendradarbiavimo aplinka, skirta neįgaliųjų emocinės būklės diagnuozei analizuoti. Neįgaliųjų diagnozės procesams formalizuoti reikia kelių metodų, kurie grindžiami skirtingais žinių vaizdavimo formalizmais, skirtingų matų parametrų atpažinimo algoritmais. Sensorinės sistemos fiksuoja pirminius stebėsenos duomenis, tačiau nenormalioms situacijos būklėms atpažinti reikia sudėtingų išvedimo metodų, taikant lanksčias neuroninių tinklų valdymo priemones. Tyrimo rezultatai pateikiami per daugelio lygmenų darbo infrastruktūrą, kuri integruoja miglota logika grindžiamų neuroninių tinklų valdymo būdus, taikant juos neįgaliojo vežimėlio valdymo konstrukcijoms, kurios leidžia valdyti vežimėlio judėjimą automatiškai valdoma trajektorija. Miglota logika grindžiamų Petri tinklų taikymas leido pademonstruoti galimybes atpažinti neįgaliojo psichologinės būsenos pokyčius ir vertinti juos laike stebint pacientus skirtingą laiką. First published online: 21 Oct 2010 Reikšminiai žodžiai: daugiaagentis sistemos valdymas, biorobotai, išskirstytosios informacinės sistemos, žinių vaizdavimo priemonės, miglota logika, neuroniniai tinklai, Petri tinklai.
... index.php? int_pageId=37, and [5]). In the proposed model of Fig. 1, two adaptive moving wheelchair-type robots are remotely communicating with two wearable human's affect sensing bio-robots. ...
Article
Full-text available
An approach is proposed in creating of an intelligent e-health care environment by modelling of an adaptive multi-agent-based e-health and e-social care system for people with movement disabilities. Human's Arousal Recognition Module (HARM) described based on online recognition of human's ECG, EDA and body temperature signals by using embedded Atmega32 type microcontrollers. Multi-agent based online motion control of multiple wheelchair-type robots is realized based on integration of an adaptive Fuzzy Neural Network Control algorithm into ATmega32 microcontroller. Human Computer Interaction in the system is realized in providing of necessary e-health care support actions for users with some movement disabilities by using Java-based JACK agent oriented environment. The dynamic multi-agent system is proposed to permanently realizing e-social care support actions for disabled by gathering data such as current position of robot and user's state information; finding decisions for given situation; sending signals to performing appropriate actions of the objects in the system. Ill. 6, bibl. 5 (in English; summaries in English, Russian and Lithuanian).
Article
This paper presents an original strategy of teaching IT curricula subjects seeking a goal of making Artificial Intelligence (AI) friendlier for the students. The strategy is based on applying learning by doing principle by in-tegration of permanent remote experimentation capabilities into adaptive e-tu-toring system. The ways of realizing of remote e-laboratory experiments are discussed based on introduction of an ambient intelligence perception within an e-laboratory. Methods of implementation of an embedded agent-based re-mote multiple object control within an e-laboratory are analyzed and some practical realization aspects are described in this paper.
Article
Full-text available
Considerable achievements in computing and telecommunication area make possible in a new way solve a wide spectrum of transportation problems, affected in the concept of intelligent transportation systems (ITS). In a technical aspect such sort systems represent a set of interacting computational nodes, equipped with various sensors, and can be treated as distributed computer systems. The distributed structure of ITS supposes parallelization of solvable transportation tasks and their distributed realization. The concept of efficiency of parallelized calculations presupposes a few aspects. Three such aspects are picked out in the work: calculation speed, efficiency of system scaling, and efficiency of parallel computations as compared to sequential ones. Proper metrics for each group are offered, allowing numerical characterization of certain aspect of parallelized computations. As such metrics the index of parallelism and speedup (PI and S), efficiency and utilization (E and U), redundancy and compression (R and C) are examined. The peculiarity of intelligent transportation systems is in, that a large volume of communications is typical for them, substantially telling on the indexes of the system. Influence of communication overheads on the total efficiency of the system is analysed in the work. The question of private indexes integration into a single integral index, characterizing the quality of ITS realization, is examined. Applicability of the suggested metrics and their evidence is illustrated by examples.
Conference Paper
This paper presents further development of intelligent multi-agent based e-health care system for people with movement disabilities. The research results present further development of multi-layered model of this system with integration of fuzzy neural control of speed of two wheelchair type robots working in real time by providing movement support for disabled individuals. An approach of filtering of skin conductance (SC) signals using Nadaraya-Watson kernel regression smoothing for emotion recognition of disabled individuals is described and implemented in the system by R software tool. The unsupervised clustering by self organizing maps (SOM) of data sample of physiological parameters extracted from SC signals was proposed in order to reduce teacher noise as well as to increase of speed and accuracy of learning process of multi-layer perceptron (MLP) training.