A Deep Learning-based approach for forecasting off-gas production and consumption in the blast furnace.

This article presents the application of a recent neural network topology known as the deep echo state network to the prediction and modeling of strongly nonlinear systems typical of the process industry. The article analyzes the results by introducing a comparison with one of the most common and efficient topologies, the long short-term memories, in order to highlight the strengths and weaknesses of a reservoir computing approach compared to one currently considered as a standard of recurrent neural network. As benchmark application, two specific processes common in the integrated steelworks are selected, with the purpose of forecasting the future energy exchanges and transformations. The procedures of training, validation and test are based on data analysis, outlier detection and reconciliation and variable selection starting from real field industrial data. The analysis of results shows the effectiveness of deep echo state networks and their strong forecasting capabilities with respect to standard recurrent methodologies both in terms of training procedures and accuracy. Supplementary Information: The online version contains supplementary material available at 10.1007/s00521-021-05984-x.

Retrospective analysis of acute HBV infections occurred in 1978-79 and 1994-95 in North-East Italy: increasing prevalence of BCP/pre-core mutants in sub-genotype D3.

BACKGROUND: At the end of the 1970s, in Italy more than 2% of the general population was HBsAg carrier. In the late '70s and late '80s, two remarkable events might have impacted on HBV strains transmitted in North-East Italy: (a) the increased HBV incidence due to parenteral drugs between 1978 and 1982; (b) the preventive anti-HIV educational campaign, started locally in 1985. METHODS: To address if those events impacted on circulating HBV variants, acute cases occurred in North-East Italy in 1978-79 (n = 50) and 1994-95 (n = 30) were retrospectively analysed. HBV sequences obtained from serum samples were subjected to phylogenetic analysis and search for BCP/pre-core and S mutations. RESULTS: HBV-D was the most prevalent genotype in both 1978-79 (43/50, 86%) and 1994-95 (24/30, 80.0%), with HBV-A in all but one remaining cases. Among HBV-D cases, sub-genotype HBV-D3 was the most prevalent (25/29, 86.2% in 1978-79; 13/16, 81.2% in 1994-95), with HBV-D1 and HBV-D2 in the remaining cases. All HBV-A cases were sub-genotype A2. Single and multiple BCP/pre-core mutations, responsible for HBeAg(-) hepatitis, were detected in 6/50 (12%) cases in 1978/79 vs. 12/30 (40.0%) in 1994/95 (p = 0.006). They were found exclusively in HBV-D; in the most abundant sub-genotype, HBV-D3, they were detected in 2/25 (8%) cases in 1978-79 vs. 6/13 (46%) in 1994-95 (p = 0.011). No vaccine escape S mutations were observed. The IDU risk factor was significantly more frequent in 1994-95 (8/30, 26.7%) than in 1978-79 (4/50, 8%) (p = 0.048). CONCLUSIONS: The above mentioned epidemiological and public health events did not affect the proportion of genotypes and sub-genotypes that remained unchanged over 16 years. In contrast, the proportion of BCP/pre-core mutants increased more than three-fold, mostly in HBV-D3, a sub-genotype highly circulating in IDUs; drug abuse likely contributed to the spread of these mutants. The findings contribute to explain a previously described major change in HBV epidemiology in Italy: the proportion of HBeAg(-) cases in the carrier cohort changed from low in late 1970s, to high at the beginning of the 2000s. In addition to other recognized factors, the increased circulation of BCP/pre-core mutants likely represents a further factor that contributed to this change.

Evaluation of rapid tests for diagnosis of acute hepatitis E.

BACKGROUND: Hepatitis E virus diagnosis still presents difficulties due to discordant results among diagnostic tests. OBJECTIVES: The aim of this study was to evaluate the performance of two rapid tests for detection of anti-HEV IgM antibodies. STUDY DESIGN: The rapid tests were compared with three commercial anti-HEV ELISA assays and one Real-Time PCR assay on 59 sera from patients with acute viral non-AC hepatitis. RESULTS: The presence of anti-HEV IgM antibodies was evaluated by two rapid tests (Wantai and Assure) on 25 HEV RNA positive samples. Anti-HEV IgM antibodies were detected in 24/25 and 23/25 samples respectively. The sensitivity and specificity of Wantai and Assure Rapid tests were evaluated using the 25 HEV RNA positive samples and 50 HEV RNA negative samples (including sera from acute-phase HAV and HBV infections and blood donors). Overall, the sensitivity of Wantai Rapid and Assure Rapid tests was 96.1% and 92.6% respectively; the specificity of the 2 tests was 100%. CONCLUSION: Our data suggest the potential use of anti-HEV IgM rapid assays as a first line test in primary health care settings, particularly useful for patients with chronic liver disease or pregnant women who urgently need an antiviral treatment.

Diagnosis of HEV infection by serological and real-time PCR assays: a study on acute non-A-C hepatitis collected from 2004 to 2010 in Italy.

BACKGROUND: The impact of hepatitis E in developed countries, like Italy, still requires a clear definition. In the present study, we evaluated HEV infection in patients with acute non-A-C hepatitis by an approach comparing data from Real-time PCR and serological assays. METHODS: In a first analysis, sera from 52 patients hospitalized with a diagnosis of acute viral non-A-C hepatitis in Italy were tested by in-house Real-Time PCR assay for identification of Hepatitis E Virus (HEV) RNA and by anti-HEV IgM and IgG assays. In a subsequent analysis, selected samples were evaluated by additional IgM tests to confirm diagnosis. RESULTS: Among the 52 samples, 21 showed positive results for all three markers (IgM, IgG and HEV RNA). One patient showed HEV RNA as single marker. Uncertain results were found in 8 samples while the remaining 22 were negative for all markers. Further analysis of the 8 undefined samples by additional IgM tests confirmed HEV infection in 1 patient. Overall, acute HEV infections were reliably identified in 23 (44.2%) out of 52 patients. CONCLUSIONS: In the present paper, we performed a study evaluating HEV infection in 52 sporadic non-A-C acute hepatitis cases. All samples were collected from 2004 to 2010 in Italy. By a diagnostic strategy based on genomic and serological assays we identified HEV infections in 23 out of 52 patients (44.2%), a percentage higher than previous estimates. Thus, the actual impact of HEV infections in Italy needs to be further evaluated on a national scale by a diagnostic strategy based on multiple and last generation assays.

Identification of low HBV-DNA levels by nucleic acid amplification test (NAT) in blood donors.

OBJECTIVE: To evaluate the presence of HBV-DNA in 22,765 consecutive blood donors, who donated blood in the period from January 2006 to August 2007 at a transfusion centre in Lazio, a region in central Italy with low HBV endemicity. METHODS: Each donation was individually tested using immunoenzymatic assays and nucleic acid amplification technologies (NAT). Samples that were reactive to generic NAT, Procleix Ultrio Assay were tested for HBV-DNA, HCV-RNA and HIV1-RNA by Discriminatory Procleix Ultrio NAT Assay. In samples that were reactive to generic NAT and negative for HBsAg, HCV-RNA and HIV1-RNA, HBV-DNA was further tested using Cobas TaqMan and an in-house nested PCR following an ultracentrifugation step. Sequence analysis confirmed HBV-DNA positivity. RESULTS: Generic NAT identified 31 (0.13%) reactive sera. HBV-DNA discriminatory NAT identified 15 positive sera; HBsAg was positive in 12 sera. Of the 5 generic NAT-reactive/discriminatory NAT-negative/HBsAg-negative sera and of the 3 HBsAg-negative/HBV-DNA discriminatory NAT-positive sera, 7 were positive to Cobas TaqMan or the in-house PCR after ultracentrifugation. The overall HBV-DNA positivity was 0.083% [19 of 22,765 donors: 12 HBsAg-positive (HBV-DNA range 10(2)-10(4) IU/mL), 7 HBsAg-negative/anti-HBc positive (HBV-DNA< 6 IU/mL)]. CONCLUSIONS: For blood transfusion safety, the significance of the finding of very low HBV-DNA levels should be further investigated. Our data indicate that in areas with a low HBV endemicity, single NAT assays may not always identify blood donations with very low HBV-DNA levels.

HCV genetic variability: from quasispecies evolution to genotype classification.

HCV is a ssRNA virus belonging to the Flaviviruses and is found worldwide worldwide in humans. Following primary infection, persistent infection develops in more than 85% of cases, which in up to 30% of cases, may progress to liver disease, cirrhosis and hepatocellular carcinoma. The virus presents a high degree of genetic variability owing to the combination of a lack of proofreading by the RNA-dependent RNA polymerase and a high level of viral replication. This genetic variability allows the classification of genotypes, subtypes, isolates and quasispecies to which epidemiological and pathogenetic significance may be associated. The features and biological implications of HCV variability and of quasispecies dynamics in infection transmission, mechanisms of chronicity and resistance to antiviral therapy are discussed.

Nucleic acid testing (NAT) for HCV RNA in Italian transfusion centres: an external quality assessment.

BACKGROUND: We conducted an external quality assessment of the results obtained in Italian transfusion centre laboratories employing nucleic acid testing (NAT) for detection of HCV RNA in donated blood. STUDY DESIGN: Of 110 transfusions centres in Italy, 101 voluntarily participated. Each laboratory received seven separate shipments of samples for HCV RNA testing by NAT. Each shipment contained 8 plasma samples for a total of 23 negative and 33 positive samples with viral loads ranging from 25 to 1000 IU/mL. RESULTS: Of the 2080 HCV RNA-negative samples, 14 (0.7%) were reported as positive. The highest percent of false-negative results (6.9%) was found on samples from the first shipment with viral loads from 75 to 100 IU/mL. In subsequent shipments, the highest false-negative percentage ranged from 0.6% for samples with viral loads of 170-1000 IU/mL to 3.4% for samples with viral loads of 35-50 IU/mL. A false-negative rate of 4.9% occurred in samples in the sixth shipment with the lowest viral load (25IU/mL). Five (4.9%) centres were identified as having laboratories with low-performance. There were no significant differences among genotypes 1b, 2c and 3a with respect to percent of false-negative results reported. CONCLUSIONS: Overall, the accuracy of NAT observed in this study of Italian transfusion centre laboratories was excellent for all HCV genotypes tested, even for samples with low HCV RNA titres.

Rare HVR1-HCV genotype 1b variants in patients with B non Hodgkin's lymphoma. Comparison with viral sequences detected in cases of lymphoproliferative disorders and B cell compartmentalisation.

We compared the E2-HVR1 region in HCV-1b positive B-NHL cases from a multicenter study with sequences from studies related to lymphoproliferative disorders and B cell compartmentalisation. We found rare and unique mutations both in B-NHL isolates and in cases with lymphoproliferative disorders and lymphocyte infection. These rare mutations could have an important effect on HVR1 region and, as a consequence, on the binding of E2 on CD81, with a possible implication for both antigenic stimulation and HCV entry. In conclusion, the HCV predominants circulating in B-NHL cases seem to be associated with clonal selection of rare variants.