Search

Intensive development of medical sciences and information technology, and their cooperation offers new opportunities and sets course of action for complex healthcare and human life protection. With the development of equipment capabilities, the emergence of new sensors and advancement of information technology, the emergence of applications supporting life and health protection, ensuring both the healthcare and human protection against various threats, can be observed. Especially in an aging society a huge demand for this type of products and services is visible, and that niche in the market is still not filled. In this paper several existing systems of remote medical care are analyzed and a concept of the constructed system of own authorship is presented.

The paper presents the use of telemedicine in intelligent supporting of otorhinolaryngologist in the diagnosis of auditory brainsteam responses. This test is easy to visualize but difficult to diagnose. The presented software system uses advanced methods of signal processing and an author's algorithm supporting the doctor by setting the characteristic points of the examination and reach the diagnosis. This paper describes the capabilities of the system and underline the benefits which are result of the nature of this application. It also highlights the benefits and opportunities introduced to this field of medicine by the described intelligent software system.

Paper presents the review of the computer aided diabetes therapy introducing GIGISim (Glucose-Insulin and Glycemic Index Web Simulator) e-learning tool based on the glucose and insulin plasma levels simulation models and genetic algorithms optimization. Together with the system, recently reported solutions assisting diabetes therapy were summarized and their functionality presented. Artificial intelligence is applied in GIGISim tools to improve patients' management and health awareness. Interactive, diabetes-dedicated simulators, supported with genetic algorithms (GA) have a great deal of educational potential for patients and their families, and may also offer a means of training for health-care professionals. It is generally believed that evolutionary algorithms (EA), which GA are a particular class of, perform consistently well across all types of optimization problems, in this case the optimization of the diabetes meal plan.

Most researchers focused on particular result ignore intermediate stages of learning process of neural networks. The unstable and transitory phenomena, discovered in neural networks during the learning process, long time after the initial stage of learning, when the network knows nothing because of random values of all weights, and long time before final stage of learning process, when the network knows (almost) everything – can be very interesting, especially when we can associate with them some psychological interpretations. Some "immature" neurons exhibit behavior that can be interpreted as source of "artificial dreams". Article presents examples of simple neural networks with capabilities which might explain the origins of dreams and myths.

Auditory brainstem response potentials (ABR) are signals calculated from the EEG signals registered as responses to an acoustic activation of the auditory system. The ABR signals provide an objective, diagnostic method, widely applied in examinations of hearing organs. The shape of the time-dependent signal, and its possible distortion, and particularly the presence or absence of characteristic waves are of great diagnostic importance. In the present work methods of automated identification of wave V are presented, using multilayer perceptron type networks. Classification methods are also proposed, based on cascade architecture of the neural networks used for recognition of the context information.

The brainstem auditory evoked potentials (BAEP) provide an objective, electrophysiological diagnostic method, widely applied in examinations of auditory organs, in particular for such cases, when the application of traditional audiometric methods is difficult or impossible, e.g. in examination of little children and infants. The shape of the time-dependent signal, and its possible distortion, and particularly the presence or absence of characteristic waves are of great diagnostic importance. In the present work methods of preliminary processing and automated identification of wave V are presented, using the self-organizing map (SOM) type neural networks for determination of the structure of the feature space and classification.

In the work excerpts of research are presented, concerning the application of modified acoustic signal processing methods in the problem of “understanding” of selected pathologies of vocal tract. The concept of the research scheme is based on the technique of advanced acoustic signal analysis and it refers to the analysis of artificial neural networks functioning in the task of recognition of selected types of vocal tract pathologies. The method is based on utilization of an internal model of the considered signal’s generator and it is directed towards such a structure analysis of the examined sound. The described method allows to achieve more subtle differentiation for signal characterized by small diversification of measurable features, observed for the classes being recognized, what is the case in the problem of identification of selected pathologies considered here.

In tasks related to the analysis and recognition of pathological speech it is often more important to provide the respective person (e.g. physician) with guidelines for a qualitative evaluation of this speech than to achieve a very accurate automated recognition. By ear it is easy to judge whether the speech is regular or deformed, but any attempt of a quantitative evaluation is not satisfactory. If the speech is transformed to a graphic form, by a proper visualization method, it is easier for a person to estimate its deformation degree by comparing the respective graphical patterns. The new visualization method proposed is based on the results obtained by application of Kohonen neural networks.

The registration of auditory brainstem response signals allows an objective analysis of processes taking place at particular levels of the neural part of the auditory system. The studies on hearing threshold using auditory brainstem response (ABR) comprise registration of a series of responses for stimuli of varying intensities and frequencies and then determination of the wave V detection threshold, which is directly correlated with the hearing threshold. The paper presents the currently realized system for analysis of ABR’s registered during screening and diagnostic examinations. Presented are the methods of preliminary processing and analysis of ABR signals, the selected space of distinctive features describing these signals and the constructed techniques for classification and automated recognition of ABR signals. The system will allow collection and distribution of both raw and processed data for conducting the research work in the field of neuroacoustics and social medicine as well as the development and testing of electromedical equipment.

Neural networks can be useful in tasks connected with analysis of auditory brainstem response (ABR) potentials, which are derived functions of bioelectric processes taking place at the brain and auditory cortex level. However, some problems have been encountered with the detection of the presence or absence of wave V in the analysed ABR potential signal. Much better results are obtained when, in addition to the primary signal of the analysed ABR, evoked for a given intensity of the acoustic signal, a context signal is used. For specification of the task, an essential problem is the input of the context signal to the neural network analysing the ABR potential. In this paper, an attempt to solve the problem is been described, and several results are shown, obtained by application of non-standard architectures for the neural networks analysing the above-mentioned signals. The neural network's architecture has quite a great influence on the results of automated classification and recognition of ABR potentials. It shows that the search for applications of neural networks with atypical architectures is worthwhile, and it proves that it is possible to build a system based on neural networks for automatically detecting the presence of wave V (with satisfactory recognition reliability) in the ABR signal. This seems to be the most essential cognitive research result of this work.

The elaboration of head-surface registration techniques for auditory potentials evoked from the brainstem (ABR) enabled the construction of objective research and diagnostic methods, which can utilized in the examinations of auditory organs. The aim of the present work was the construction of a method, making use of the neural network techniques, enabling an automated detection of wave V in the ABR signals. The basic problem encountered in any attempts of automated analysis of the auditory potentials is connected with impossibility of a reliable evaluation of a single response evoked by a weak acoustic signal. It has been assumed that considerably better detection results should be obtained, when additional context information will be provided to the network's input. This assumption has been verified using complex, hybrid neural networks. As a result about 90% of correct recognitions has been achieved.

In many problems of medical diagnosis as well as therapy and rehabilitation, evaluation of the deformed speech signal quality is required. In problems of distorted speech diagnosis the regular methods of speech signal preprocessing and classification, used in speech or voice recognition, totally fail. Also the standard speech signal parametrization techniques (e.g. LPC or cepstral coefficients) cannot satisfactorily describe pathological speech because of its dissimilar phonetic and acoustic structure compared with correct speech, and also because the aim of the recognition process is totally different. In the paper a new method for processing and classification of pathologically deformed speech, based on neural networks techniques, is presented and discussed.

In the present work excerpts of research are presented, concerning the application of modified acoustic signal processing methods in the problem of "understanding" of selected pathologies of vocal tract. The presented concept of the research scheme is based on the technique of advanced acoustic signal analysis and it refers to the analysis of artificial neural networks functioning in the task of recognition of selected types of vocal tract pathologies. It is recommended here that the simple process of signal recognition should be replaced by a more advanced method or its analysis, called the process of automated understanding of the signal. The method is based on utilization of an internal model of the considered signal's generator and it is directed towards such a structure analysis of the examined sound, which enables its identification as a result of cognitive resonance. The described method allows to achieve more subtle differentiation for signal characterized by small diversification of measurable features, observed for the classes being recognized, what is the case in the problem or identification of selected pathologies considered here. The circumstances mentioned above suggest a consideration of more knowledge-based approach to the discrimination of acoustic signals, labeled here as a technique of signal understanding.