117,482 research outputs found

    A deep Q-learning based approach applied to the snake game

    No full text
    In recent years, one of the highest challenges in the field of artificial intelligence has been the creation of systems capable of learning how to play classic games. This paper presents a Deep Q-Learning based approach for playing the Snake game. All the elements of the related Reinforcement Learning framework are defined. Numerical simulations for both the training and the testing phases are presented. A particular focus is given to the associated Neural Network hyperparameters tuning, which is a crucial step in the agent design process and for the achievement of a desired target level of performance

    sj-docx-1-sph-10.1177_19417381211070834 – Supplemental material for Effects of Blood Flow Restriction Training on Clinical Outcomes for Patients with ACL Reconstruction: A Systematic Review

    No full text
    Supplemental material, sj-docx-1-sph-10.1177_19417381211070834 for Effects of Blood Flow Restriction Training on Clinical Outcomes for Patients with ACL Reconstruction: A Systematic Review by Mark Colapietro, Brandon Portnoff, Sayers John Miller, Wayne Sebastianelli and Giampietro L Vairo in Sports Health: A Multidisciplinary Approach</p

    Limitazione della responsabilità sanitaria in tempi di emergenza da COVID-19

    No full text
    [Abstract non presente. Si riporta l\u27inizio dell\u27editoriale]Nell’odierno contesto storico, gravato da un’emergenza sanitaria senza precedenti, ci si domanda sotto quali forme la responsabilità medica debba essere inquadrata e quali siano le singole responsabilità che possano essere addebitate all’esercente la professione sanitaria, di cui all’art. 7 della legge 8 marzo 2017, n. 24, e alla struttura sanitaria..

    Perioperative complications of sleeve gastrectomy: Review of the literature

    No full text
    Sleeve gastrectomy (SG) has known a spectacular rise worldwide during the last decade. The absence of digestive anastomosis simplifies the surgical technique, reducing anastomosis-related complications such as fistula, stricture and marginal ulcer. Furthermore, the respect for digestive continuity preserves the functions of pylorus, that regulates gastric emptying, and duodenum, where calcium, B vitamins and iron are absorbed. Despite the multiple advantages, SG also has specific complications such as bleeding, stenosis, portal thrombosis and leak. The staple line leak at the oesophagogastric junction is the most feared complication and its prevention remains difficult, as the involved mechanisms have been only partially elucidated. Its management is long and requires a multidisciplinary technical platform including Intensive Care Unit, digestive endoscopy and interventional radiology as well as a specialised surgeon. The aim of this review is to explain in detail the perioperative complications of SG, their prevention and treatment, referring to the most recent available literature

    Assessment of the watershed DEM mesh size influence on a large dam design hydrograph

    No full text
    The present study aims to assess the digital elevation model (DEM) resolution influence on the peak flow estimation for the design hydrograph of a large dam. This was executed by comparing the design hydrograph peak flows, with respect to a 2000years return period, which wereestimated for the Pietrarossa dam, in the South-East of Sicily, for different DEM spatial resolutions. The methodology consisted of the watershed extraction from the catchment basin in which the directly wired area belongs. Furthermore, the intensity duration frequency (IDF) curves were estimated starting from the observational time seriescollected by two rain gauges located near the dam. Finally, through a rainfall-runoff transformation, the design hydrographs were obtained by using both the watershed and the IDF curves. Considering different spatial resolutions, it was found that both the peak flow and the total volume decreases as the DEM spatial resolution decreases

    On precipitation measurements collected by a weather radar and a rain gauge network

    No full text
    The present work aims to analyze the correspondence between the rainfall radar estimates and the rain gauges measurements collected at different distances from radar. The radar data utilized in this work have been collected by the Polar 55C weather radar, located in Roma Tor Vergata research area, from 2008 to 2009. The scanning strategy adopted by Polar 55C prefigured the cyclical repetition of eight sweeps in all directions with constant elevation. Each 5 minutes 8 Plan Position Indicators are acquired, each one with a different elevation angle, ranging from 0.5 to 7.5_. This study considers measurements collected at 1.5_ elevation. The noise level is determined by supposing that at great distance the sampling volume is likely situated in an atmospheric region above the precipitation. In this way the modal value in the last two range-bins has been chosen as a reference to determine the noise level at the receiver. The range-bins whose reflectivity doesn’t exceed noise level were considered affected by noise. The modality developed for the ground clutter removal is based on the existence of typical values for the standard deviations of the differential reflectivity and of the differential phase when the radar return is caused by precipitation. In fact in the presence of meteorological echoes at the receiver, these standard deviations can be expressed by the width of the Doppler spectrum and the co-polar correlation coefficient, about which is well-known the variation range characteristic of rainfall. Only the radar reflectivity which corresponds to meteorological signal was converted into rainfall intensity by using a parametric algorithm. Finally, the radar rainfall intensity values were remapped onto a 1 square kilometre Cartesian grid, by assigning to each pixel the mean of the rainfall values estimated in the radar range-bins belonging to the pixel. Rain gauges located at different distances from Polar 55C were selected so that most of ranges in the scanning circle are covered. Moreover only rain gauges not placed in areas where the radar beam is blocked were considered. The rain gauges data were compared with rainfall radar estimates in the pixels where are located the rain gauges considered. The ratio G/R between rainfall amounts rain gauges measurements and rainfall amounts radar estimates was calculated against the distance from radar, by considering all the events utilized in this work. A trend was found; the greater the distance from the radar, the higher the ratio G/R. Once the trend has been found, a best fitting line was used to find the radar error at a given range and the radar rainfall estimates were consequently corrected

    Persistence ed effect of rotenone on oil quality in two Italian olive cultivars

    No full text
    The aims of this work are to assess the persistence of rotenone in oil and drupes of olive plants of cultivars Nocellara del Belice (NB) and Cassanese (CA), and to compare the quality of oils from rotenone-treated and untreated plants. Samples of drupes and oil were analyzed at 2, 12, 22 and 30 days after treatment. Rotenone levels in drupes of treated plants declined by about 50% after 22 days from treatment (0.037 mg kg-1 in NB and 0.039 mg kg-1 in CA), whereas the respective values in the oil were higher (0.209 mg kg-1 in NB and 0.229 mg kg-1 in CA) and had a lower decay half-life (4.02 days in NB and 4.71 days in CA). For both cultivars, no significant differences in oil physicochemical and nutritional parameters were found between the two treatments. The panel test of oils extracted after 22 days did not reveal significant differences in unpleasant aromatic notes nor defects between the two treatments. Our results confirm that serious doubts remain about the safety and healthiness of oils extracted from drupes treated with rotenone. This information could assess the real risk in the use of this product for plant protection in olive growing
    corecore