surface temperature is projected to rise throughout the 21st century under all assessed emission scenarios [2]. Such global warming directly affects precipitations because the water holding capacity of air increases by about 7% per degree C [3] that leads to more water vapor being retained in the atmosphere. Storms, thunderstorms, extra-tropical rains, snow, are therefore supplied with more moisture and produce more extreme precipitation events. Such events are observed to be widely occurring, even where total precipitation is decreasing, and, in combination with rapid snow melting, they increases the risk of flooding. Given that floods are usually weather-induced, meteorological services provide local authorities with a periodical weather and flood hazard forecast that contains an encoded alert level on a predetermined set of geographical areas. The alert level is used to trigger actions according to a predefined

Abstract-Floods are major natural disasters which cause deaths and material damages every year. Monitoring these events is crucial in order to reduce both the affected people and the economic losses. In this work we train and test three different Deep Learning segmentation algorithms to estimate the water area from river images, and compare their performances. We discuss the implementation of a novel data chain aimed to monitor river water levels by automatically process data collected from surveillance cameras, and to give alerts in case of high increases of the water level or flooding. We also create and openly publish the first image dataset for river water segmentation.

The detection of speci®c IgM antibodies has been the most frequently used serological marker for diagnosing recent toxoplasmosis. However, the persistence of speci®c IgM antibodies in some patients and the use of tests with a low speci®city have complicated the interpretation of serological results when toxoplasmosis is suspected. The purpose of the present study was to determine the value of newer serological techniques in the diagnosis of acute acquired toxoplasmosis. Sixty-four sera, 31 from patients with Toxoplasma gondii infection and 33 from patients with latent infection, were tested. Anti-T. gondii IgA was measured by two antibody capture ELISA tests (Platelia 1 Toxo IgA and ETI-TOXOK A) and an automated direct ELISA (IMx 1 Toxo IgA); all three assays detected antibody levels compatible with a recent infection in sera from all 31 patients with acute toxoplasmosis. However, signi®cant levels of IgA were also detected with high frequency by all three assays in sera from patients with latent infection. IgE antibodies detected by IgE immunosorbent agglutination assay (ISAGA) were present in 26 (84%) of 31 patients with acute toxoplasmosis and in sera from two subjects with latent infection taken >1 year after the beginning of the clinical symptoms of infection. Thirty (97%) of 31 patients with a recent T. gondii infection and 15 (45%) of 33 subjects with latent infection had an AC/HS pattern compatible with acute toxoplasmosis. The avidity of T. gondii IgG was evaluated by two methods. One method was based on the titration of each serum sample and calculation of the titres, in the absence and presence of urea, in relation to a de®ned cut-off value. In the other method, a single serum dilution was used and the absorbances of the reactions in the presence and absence of urea were compared. The titration method was more sensitive for diagnosing recent primary infection; all 31 sera from patients with acute toxoplasmosis had avidity indices compatible with acute toxoplasmosis by the titration method, whereas with the single dilution method, sera from four patients had equivocal results. In the 33 individuals with latent infection, similar results were obtained with the two avidity methods; only one serum sample had a non-compatible avidity value with the titration method. The results obtained in the present study show that the current serological markers used for diagnosing acute acquired toxoplasmosis have signi®cant limitations. The data suggest that determination of the avidity of T. gondii-speci®c IgG by the titration method in patients with detectable IgM antibodies de®nes most accurately the stage of infection by T. gondii.

Polystyrene latex microspheres are efficient supports for immunological reactions. In the presence of suitable functional groups these microspheres may react by covalent binding with antigens or antibodies and have been extensively used in immunoassay technique. By using the emulsion polymerization and diazotization methods, a kind of functionalized latex was prepared. The preparation of this latex consisted of six syntheses of polystyrene followed by nitration, amination and diazotization. A narrow distribution of particle size was obtained in all the cases. The synthesis that presented 98% of monomer conversion and the largest particle size diameter 0.2 ± 0.05 μm was chosen for the subsequent chemical treatments. The chemical characterizations of every step were performed by Fourier transform infrared and Raman spectroscopies. The particle size was determined by a submicrometer particle analyzer and scanning electron microscopy. The polydiazostyrene latex obtained was sensitized using bovine serum albumin (BSA) and when it was submitted to agglutination assay in the presence of a rabbit serum containing specific antibodies to BSA demonstrated efficient results. © 1997 John Wiley & Sons, Ltd.

Immunodiffusion (ID) is the serologic test most frequently used for the diagnosis and posttherapy follow-up of patients with paracoccidioidomycosis (PCM). The ID test is highly specific (100%), but its sensitivity is relatively low (90%), leading to false-negative results. The aim of this study was to determine the profiles of antibodies in sera from patients with proven PCM and with negative results in the ID test (ID neg ) versus positive results in the ID test (ID pos ). We analyzed 46 sera from patients with active PCM for total immunoglobulin G (IgG) and IgG subclass responses to Paracoccidioides brasiliensis gp43 antigen (treated or not treated with sodium metaperiodate) by enzyme-linked immunosorbent assay and immunoblotting. Immunoblotting showed that both ID neg and ID pos sera recognized predominantly the gp43 fraction of the P. brasiliensis antigen used in the ID test. ID neg sera contain low-avidity antibodies, low levels of specific IgG (total) and IgG1, and high levels of IgG2 compared with ID pos sera. The antibodies present in ID neg sera were predominantly directed against carbohydrate epitopes, since treatment with sodium metaperiodate resulted in a significant decrease in antibody reactivity. These data suggest that the lack of reactivity of sera from PCM patients in the ID test may be related to the production of low-avidity IgG2 antibodies directed against carbohydrate epitopes.