Evaluation of free libraries for visual recognition of art imagery on mobile devices

This paper describes a systematic research about free software solutions and techniques for art imagery computer recognition problem.

Changes in patterns of practice for prostate cancer radiotherapy in Italy 1995-2003. A survey of the Prostate Cancer Study Group of the Italian Radiation Oncology Society.

Aims and background. In 2002, a survey including 1759 patients treated from 1980 to 1998 established a "benchmark" Italian data source for prostate cancer radiotherapy. This report updates the previous one. Methods. Data on clinical management and outcomes of 3001 patients treated in 15 centers from 1999 through 2003 were analyzed and compared with those of the previous survey. Results. Significant differences in clinical management (-10% had abdominal ma-gnetic resonance imaging; +26% received ≥70 Gy, +48% conformal radiotherapy, -20% pelvic radiotherapy) and in G3-4 toxicity rates (-3.8%) were recorded. Actuarial 5-year overall, disease-specific, clinical relapse-free, and biochemical relapse-free survival rates were 88%, 96%, 96% and 88%, respectively. At multivariate analysis, D'Amico risk categories significantly impacted on all the outcomes; higher radiotherapy doses were significantly related with better overall survival rates, and a similar trend was evident for disease-specific and biochemical relapse-free survival; cumulative probability of 5-year late G1-4 toxicity was 24.8% and was significantly related to higher radiotherapy doses (P <0.001). Conclusions. The changing patterns of practice described seem related to an improvement in efficacy and safety of radiotherapy for prostate cancer. However, the impact of the new radiotherapy techniques should be prospectively evaluated.

Greenberg i Danto : De l'últim gran relat a la fi de l'art

A partir de l'anàlisi comparatiu de l'obra de dos grans teòrics de la modernitat i la postmodernitat com són Clement Greenberg i Arthur C. Danto es fa un estudi de com aquests dos autors han arribat a construir els seus discursos i, en relació a això, com ha anat evolucionant el concepte d'art de la modernitat a la postmodernitat. La qüestió central que se'ns planteja és: què ha canviat en la idea d'art de la postmodernitat respecte a la idea o a la concepció d'art anterior, per arribar-nos a qüestionar si realment hi ha art i, si n'hi ha, què és art i que no ho és? Existeix una idea universal de què ha de ser art? O aquesta és una qüestió intrínsicament relacionada amb la societat en què es produeix; en aquest cas, la societat occidental?

Art painting Data Collection

The classification of Art painting images is a computer vision applications that isgrowing considerably. The goal of this technology, is to classify an art paintingimage automatically, in terms of artistic style, technique used, or its author. For thispurpose, the image is analyzed extracting some visual features. Many articlesrelated with these problems have been issued, but in general the proposed solutionsare focused in a very specific field. In particular, algorithms are tested using imagesat different resolutions, acquired under different illumination conditions. Thatmakes complicate the performance comparison of the different methods. In thiscontext, it will be very interesting to construct a public art image database, in orderto compare all the existing algorithms under the same conditions. This paperpresents a large art image database, with their corresponding labels according to thefollowing characteristics: title, author, style and technique. Furthermore, a tool thatmanages this database have been developed, and it can be used to extract differentvisual features for any selected image. This data can be exported to a file in CSVformat, allowing researchers to analyze the data with other tools. During the datacollection, the tool stores the elapsed time in the calculation. Thus, this tool alsoallows to compare the efficiency, in computation time, of different mathematicalprocedures for extracting image data.