Simplification of a shape of a coastline is one of the best-described issues of quantitative generalisation. Schematisation of a coastline shape is a process, which may be relatively easily described by means of an algorithmic formula. However, the majority of algorithms consider only geometric aspects and river and road networks are generalised by means of the same parameters. Many described methods of direct transfer of subjective ways of manual generalisation to computer systems have turned out to be ineffective. Application of fractal analysis is an attempt aiming at objective implementation of a process of automated cartographic generalisation by means of selection of parameters of algorithms of simplification of lines, preceded by analysis of local geometric features of modelled objects. The, so-called mono-fractal dimension of objects, commonly used in cartometric analysis, DJ, specifies the averaged level of filling of available space only. The multi-fractal dimension of analysed objects, as, for example of a coastline, determined by means of a method proposed by the author, specifies the multi-fractal spectrum of dimensions, D(q). The range of obtained values of the parameter DJ( l .05-e- 1.42) allows for assumption that the coastline has multi-fractal properties. In this paper the author proposes development of new descriptive and research tools, which may be used for investigation of local geometric features of objects presented on a map, as well as for simplification of shapes of objects in the process of cartographic generalisation.

W artykule omówiono wybrane współczesne metody wyznaczania podstawowych jednostek geomorfologicznych w oparciu o analizy morfometryczne numerycznego modelu rzeźby terenu. Przedstawiono także metody klasyfikowania danych wielocechowych oraz pokazano możliwość ich wykorzystania w kartografii i geomorfologii. Szczególną uwagę zwrócono na możliwość zastosowania sieci neuronowych, a zwłaszcza uczonych w trybie ,,bez nauczyciela" sieci Kohonena jako narzędzia nienadzorowanej klasyfikacji i uogólniania danych przestrzennych. Istnieje wiele klasycznych kryteriów i wskaźników morfometrycznych stosowanych do klasyfikacji typologicznej form powierzchni terenu takich jak nachylenie zboczy, deniwelacje lokalne, wskaźnik krętości, wskaźnik zwartości/rozczłonkowania itp. W przeprowadzonych badaniach zastosowano zestaw prostych kryteriów morfometrycznych, stosując do ich analizy sieci neuronowe Kohonena. Uzyskane wyniki wskazują, iż sztuczne sieci neuronowe uczone w trybie ,,bez nauczyciela" mogą być wykorzystywane jako narzędzie nienadzorowanej klasyfikacji wielocechowych danych przestrzennych. Zastosowanie algorytmu Kohonena do analizy rzeźby terenu (reprezentowanej przez NMT) umożliwia wydzielenie podstawowych struktur geomorfologicznych. Dla uzyskanych wyników istotny jest nie tylko dobór danych wejściowych (wskaźników morfometrycznych), lecz także rozmiar warstwy wyjściowej sieci Kohonena. Zastosowanie mniejszej liczby neuronów w tej warstwie pozwala na większy stopień uogólnienia wyników i zwiększenie rozmiarów wydzielonych jednostek terenowych. Istotny wpływ na przebieg iteracyjnego procesu klasyfikacji ma także liczba epok obliczeniowych. Użycie zbyt malej liczby iteracji powoduje ,,niedouczenie" sieci neuronowej i w konsekwencji uzyskanie obrazu struktur niespójnych przestrzennie. Interesujące jest także wykorzystanie zaproponowanych algorytmów do generalizacji numerycznego modelu rzeźby terenu. Porównanie uzyskanych wyników z generalizacją manualną lub uśrednieniem modelu podstawowego prowadzi do wniosku, że algorytm Kohonena może być wykorzystywany jako alternatywna metoda uogólniania modelu rzeźby

The issue of line simplification is one of the fundamental problems of generalisation of geographical information, and the proper parameterisation of simplification algorithms is essential for the correctness and cartographic quality of the results. The authors of this study have attempted to apply computational intelligence methods in order to create a cartographic knowledge base that would allow for non-standard parameterisation of WEA (Weighted Effective Area) simplification algorithm. The aim of the conducted research was to obtain two independent methods of non-linear weighting of multi-dimensional regression function that determines the “importance” of specific points on the line and their comparison to each other. The first proposed approach consisted in the preparation of a set of cartographically correct examples constituting a basis for teaching a neural network, while the other one consisted in defining inference rules using fuzzy logic. The obtained results demonstrate that both methods have great potential, although the proposed solutions require detailed parameterisation taking into account the specificity of geometric variety of the source data.

The brain is subject to damage, due to ageing, physiological processes and/or disease. Some of the damage is acute in nature, such as strokes; some is more subtle, like white matter lesions. White matter lesions or hyperintensities (WMH) can be one of the first signs of micro brain damage. We implemented the Acoustocerebrography (ACG) as an easy to use method designed to capture differing states of human brain tissue and the respective changes.

Aim: The purpose of the study is to compare the efficacy of ACG and Magnetic Resonance Imaging (MRI) to detect WMH in patients with clinically silent atrial fibrillation (AF).

Methods and results: The study included 97 patients (age 66.26 ± 6.54 years) with AF. CHA2DS2-VASc score (2.5 ±1.3) and HAS BLED (1.65 ± 0.9). According to MRI data, the patients were assigned into four groups depending on the number of lesions: L0 – 0 to 4 lesions, L5 – 5 to 9 lesions, L10 – 10 to 29 lesions, and L30 – 30 or more lesions. Authors found that the ACG method clearly differentiates the groups L0 (with 0–4 lesions) and L30 (with more than 30 lesions) of WMH patients. Fisher’s Exact Test shows that this correlation is highly significant (p < 0:001).

Conclusion: ACG is a new, easy and cost-effective method for detecting WMH in patients with atrial fibrillation

The main objective of this study is to develop an echocardiographic model of the left ventricular and numerical modeling of the speckles- markers tracking in the ultrasound (ultrasonographic) imaging of the left ventricle. The work is aimed at the creation of controlled and mobile environment that enables to examine the relationships between left ventricular wall deformations and visualizations of these states in the form of echocardiographic imaging and relations between the dynamically changing distributions of tissue markers of studied structures.

Objectives: In the article we describe the new, high frequency, 20 MHz scanning/Doppler probe designed to measure the flow mediated dilation (FMD) and shear rate (SR) close to the radial artery wall.

Methods: We compare two US scanning systems, standard vascular modality working below 12 MHz and high frequency 20 MHz system designed for FMD and SR measurements. Axial resolutions of both systems were compared by imaging of two closely spaced food plastic foils immersed in water and by measuring systolic/diastolic diameter changes in the radial artery. The sensitivities of Doppler modalities were also determined. The diagnostic potential of a high frequency system in measurements of FMD and SR was studied in vivo, in two groups of subjects, 12 healthy volunteers and 14 patients with stable coronary artery disease (CAD).

Results: Over three times better axial resolution was demonstrated for a high frequency system. Also, the sensitivity of the external single transducer 20 MHz pulse Doppler proved to be over 20 dB better (in terms of a signal-to-noise ratio) than the pulse Doppler incorporated into the linear array. Statistically significant differences in FMD and FMD/SR values for healthy volunteers and CAD patients were confirmed, p-values < 0:05. The areas under Receiver Operating Characteristic (ROC) curves for FMD and FMD/SR for the prediction CAD had the values of 0.99 and 0.97, respectively.

Conclusions: These results justify the usefulness of the designed high-frequency scanning system to determine the FMD and SR in the radial artery as predictors of coronary arterial disease.

The wide access to source data, published by numerous websites, results in situation, when information acquisition is not a problem any more. The real problem is how to transform information in the useful knowledge . Cartographic method of research, dealing with spatial data, has been serving this purpose for many years. Nowadays, it allows conducting analyses at the high complexity level, thanks to the intense development in IT technologies, The vast majority of analytic methods utilizing the so-called data mining and data enrichment techniques, however, concerns non-spatial data. According to the Authors, utilizing those techniques in spatial data analysis (including analysis based on statistical data with spatial reference), would allow the evolution of the Spatial Information Infrastructure (SII) into the Spatial Knowledge Infrastructure (SKI). The SKI development would benefit from the existence of statistical geoportal. Its proposed functionality, consisting of data analysis as well as visualization, is outlined in the article. The examples of geostatistical analyses (ANOVA and the regression model considering the spatial neighborhood), possible to implement in such portal and allowing to produce the “cartographic added value”, are also presented here