Indoor position estimation has become an attractive research topic due to growing interest in location-aware services. Nevertheless, satisfying solutions have not been found with the considerations of both accuracy and system complexity. From the perspective of lightweight mobile devices, they are extremely important characteristics, because both the processor power and energy availability are limited. Hence, an indoor localization system with high computational complexity can cause complete battery drain within a few hours. In our research, we use a data mining technique named boosting to develop a localization system based on multiple weighted decision trees to predict the device location, since it has high accuracy and low computational complexity. The localization system is built using a dataset from sensor fusion, which combines the strength of radio signals from different wireless local area network access points and device orientation information from a digital compass built-in mobile device, so that extra sensors are unnecessary. Experimental results indicate that the proposed system leads to substantial improvements on computational complexity over the widely-used traditional fingerprinting methods, and it has a better accuracy than they have.

PurposeTo develop a digital image processing method to quantify structural components (smooth muscle fibers and extracellular matrix) in the vessel wall stained with Masson’s trichrome, and a statistical method suitable for small sample sizes to analyze the results previously obtained.MethodsThe quantification method comprises two stages. The pre-processing stage improves tissue image appearance and the vessel wall area is delimited. In the feature extraction stage, the vessel wall components are segmented by grouping pixels with a similar color. The area of each component is calculated by normalizing the number of pixels of each group by the vessel wall area. Statistical analyses are implemented by permutation tests, based on resampling without replacement from the set of the observed data to obtain a sampling distribution of an estimator. The implementation can be parallelized on a multicore machine to reduce execution time.ResultsThe methods have been tested on 48 vessel wall samples of the internal saphenous vein stained with Masson’s trichrome. The results show that the segmented areas are consistent with the perception of a team of doctors and demonstrate good correlation between the expert judgments and the measured parameters for evaluating vessel wall changes.ConclusionThe proposed methodology offers a powerful tool to quantify some components of the vessel wall. It is more objective, sensitive and accurate than the biochemical and qualitative methods traditionally used. The permutation tests are suitable statistical techniques to analyze the numerical measurements obtained when the underlying assumptions of the other statistical techniques are not met.

Research based on indoor location systems has recently been developed due to growing interest in locationaware services to be implemented in light mobile devices. Most of this work is based on received signal strength (RSS) from access points. However, a major drawback from using RSS is its variability due to indoor multipath effect caused by reflection, diffraction and scattering of signal propagation. Therefore, different device orientations in a fixed location provide significant and different RSS values. In this paper, we propose to extend fingerprinting with device orientation information. Implementation of our location system is based on data mining techniques employing decision tree algorithms. Experimental results demonstrate that using RSS samples with the device orientation information improves the location estimation with high accuracy.

WHAT THIS PAPER ADDSThis observational research demonstrates quantitatively that near infrared illumination increases the visualisation of the subcutaneous venous network between 2.6 and 16.2 times vs. the traditional system of visual inspection. Morphological details of the superficial venous network of the lower extremities are more clearly displayed than with the naked eye, which allows for future studies to be designed to obtain better knowledge of the superficial venous network. This could be of help in understanding the behaviour in both control limbs (CEAP C 0A class) and limbs with venous incompetence (classes C 0S and C 1 of the CEAP classification).Objective: The subcutaneous venous network (SVN) is difficult to see with the naked eye. Near infrared illumination (NIr-I) claims to improve this. The aims of this observational study were to investigate whether there are differences between the different methods; to quantify the length and diameter of SVNs; and to confirm if they differ between C 0A and C 1 CEAP limbs. Methods: In total, 4 796 images, half of them from the visible spectrum (VS) and the other half from the nearninfrared spectrum (NIrS), belonging to 109 females (C 0A : n ¼ 50; C 1 CEAP: n ¼ 59) were used to establish the morphological characteristics of the SVN by visual analysis. With Photoshop CS4, SVN diameters and lengths were obtained by digital analysis of 3 052 images, once the images of whole extremities were excluded. Results: On NIr-I, the diameters, trajectories, and colouration of SVNs of C 1 limbs appeared more irregular than SVNs of C 0A limbs. Compared with the VS images, NIr-I allowed visualisation of a greater length of the SVN in both groups (p < .010). This capacity varied from 2.6 AE 0.9 times (C 1 ) to 16.2 AE 11.9 (C 0A ). While the SVN length seen in the VS images from C 1 limbs was greater than observed in C 0A limbs (p < .001), differences between NIr-I images only existed in the lateral part of the lower leg (p ¼ .016). With NIr-I, the median diameter of the C 1 CEAP SVN veins was 5.8 mm (interquartile range [IQR] 4.3e7.5 mm), while the median diameter in C 0A SVN limbs was 2.6 mm (IQR 2.0e3.6 mm) (p < .001). Conclusion:The NIr-I reveals the characteristics of the SVN better than the naked eye. Further studies are required to determine the significance of the changes in the SVN in C 0A and C 1 limbs, and the factors causing them.

Resumen: La clasificación de textos, en entornos en los que el volumen de datos a clasificar es tan elevado que resulta muy costosa la realización de esta tarea por parte de humanos, requiere la utilización de clasificadores de textos en lenguaje natural automáticos. El clasificador propuesto en el presente estudio toma como base la Wikipedia para la creación del corpus que define una categoría mediante técnicas de Procesado de Lenguaje Natural (PLN) que analizan sintácticamente los textos a clasificar. El resultado final del sistema propuesto presenta un alto porcentaje de acierto, incluso cuando se compara con los resultados obtenidos con técnicas alternativas de Aprendizaje Automático.Palabras clave: Categorización de textos; Wikipedia; tf-idf; Aprendizaje Automático; Procesado de Lenguaje Natural.Abstract: Automatic Text Classifiers are needed in environments where the amount of data to handle is so high that human classification would be ineffective. In our study, the proposed classifier takes advantage of the Wikipedia to generate the corpus defining each category. The text is then analyzed syntactically using Natural Language Processing software. The proposed classifier is highly accurate and outperforms Machine Learning trained classifiers.