Publications

Boron efficiently absorbs neutrons due to its large cross section. Thus, boron containing materials are an effective shield to neutrons and are commonly used as containment barriers in nuclear reactors. The most economical way to include boron into shielding structures is to prepare B-rich mortars or concretes, to be used as structural elements or as plastering. However, colemanite [Ca(B3O4(OH)3)(H2O)], the most abundant B-containing mineral, is sufficiently soluble to release enough borate ions in solution to indefinitely stop Portland cement hydration. Here we present the formulation of hydraulically active binders containing 50% of colemanite. They are based on blends of calcium aluminate cements and blastfurnace slag. The main hydration product in the absence of colemanite is strätlingite along with other AFm phases. MgO causes an increasing hydrotalcite precipitation, and fly ashes further increase strätlingite content. The presence of colemanite causes the precipitation of B-ettringite, where B(OH)4¯ ions substitute for sulphate ions. These binders set in one day and harden in 4 days. The addition of hydrated lime in the formulations brings about the additional precipitation of B-containing AFm phases, where the trigonal HBO32- ion constitutes the interlayer between positive [Ca2Al(OH)6]+ sheets. These binders set in few hours and harden in one day.

Autonomous vehicles (AVs) must share space with pedestrians, both in carriageway cases such as cars at pedestrian crossings and off-carriageway cases such as delivery vehicles navigating through crowds on pedestrianized high-streets. Unlike static obstacles, pedestrians are active agents with complex, interactive motions. Planning AV actions in the presence of pedestrians thus requires modelling of their probable future behavior as well as detecting and tracking them. This narrative review article is Part I of a pair, together surveying the current technology stack involved in this process, organising recent research into a hierarchical taxonomy ranging from low-level image detection to high-level psychology models, from the perspective of an AV designer. This self-contained Part I covers the lower levels of this stack, from sensing, through detection and recognition, up to tracking of pedestrians. Technologies at these levels are found to be mature and available as foundations for use in high-level systems, such as behavior modelling, prediction and interaction control.

Conducting “manual” transcriptions and analyses is unsustainable for most historical oral archives because they require a remarkable amount of funds and time. The FONTI 4.0 project aims at exploring the suitability of automatic transcription and information extraction technologies for making historical oral sources available. In this work, we conducted an experiment to test the performance of two commercial speech-to-text services (Google Cloud Speech-to-text and Amazon Transcribe) on digitized oral sources. We created an eight-hour corpus made of manually transcribed and annotated historical speech recordings in TEI format. The results clearly show how audio quality and disturbing elements (e.g., overlaps, foreign words, etc.) impact on the automatic transcription, showing what needs to be improved for implementing an unsupervised transcription chain.

This paper presents a workflow and novel digital filters for compensating speed and equalization errors that can impact digitized audio open-reel tapes. We examine three frequent cases of mismatch between recording and reproducing standards: NAB 3.75 ips – CCIR 7.5 ips; NAB 3.75 ips – CCIR 15 ips; NAB 7.5 ips – CCIR 15 ips. Three MUSHRA-inspired tests (“sets”) containing ≥ 21 participants were used to perceptually assess the workflow and digital filters, using excerpts of music and voice. The results indicated that the digital correction filters performed well, although the electroacoustic stimuli in Set C provided mixed results, suggesting that the style of the music used in perception tests should not be overlooked.

Background: Central venous catheter (CVC) placement is a routine procedure but is potentially associated with severe complications. Relatively small studies investigated if the use of ultrasound is effective in bridging the skill gap between proficient and not proficient operators, while patient safety during training remains a controversial topic. The first aim of this study was to evaluate if resident proficiency affects the failure rate in CVC positioning under ultrasound guidance. In addition, it aimed to investigate the different rate of complications between proficient and non proficient residents. Methods: We conducted a cohort study including CVC placed by residents at the University Hospital of Padova, from November 1, 2012 to July 9, 2020 comparing proficient and non proficient residents. To avoid bias the two cohorts were matched using propensity score. Results: A total of 356 residents positioned 2310 CVC during the 8 year study period. Among them, two groups of 1060 CVCs each were matched with a propensity score analysis. There was no difference in the failure rate among the groups (2.8 vs 2.7%, p-value 0.895). Moreover, cohorts had the same rate of hematomas, catheter tip malposition, arterial puncture and pneumothorax. No cases of hemothorax were reported. Conclusions: We found the same rate of success and incidence of adverse complications among cohorts, meaning that the process of skill acquisition is safe as long as appropriate training and direct supervision by a senior consultant are available.

Background: Pathophysiological features of coronavirus disease 2019-associated acute respiratory distress syndrome (COVID-19 ARDS) were indicated to be somewhat different from those described in nonCOVID-19 ARDS, because of relatively preserved compliance of the respiratory system despite marked hypoxemia. We aim ascertaining whether respiratory system static compliance (Crs), driving pressure (DP), and tidal volume normalized for ideal body weight (VT/kg IBW) at the 1st day of controlled mechanical ventilation are associated with intensive care unit (ICU) mortality in COVID-19 ARDS. Methods: Observational multicenter cohort study. All consecutive COVID-19 adult patients admitted to 25 ICUs belonging to the COVID-19 VENETO ICU network (February 28th–April 28th, 2020), who received controlled mechanical ventilation, were screened. Only patients fulfilling ARDS criteria and with complete records of Crs, DP and VT/kg IBW within the 1st day of controlled mechanical ventilation were included. Crs, DP and VT/kg IBW were collected in sedated, paralyzed and supine patients. Results: A total of 704 COVID-19 patients were screened and 241 enrolled. Seventy-one patients (29%) died in ICU. The logistic regression analysis showed that: (1) Crs was not linearly associated with ICU mortality (p value for nonlinearity = 0.01), with a greater risk of death for values < 48 ml/cmH2O; (2) the association between DP and ICU mortality was linear (p value for nonlinearity = 0.68), and increasing DP from 10 to 14 cmH2O caused significant higher odds of in-ICU death (OR 1.45, 95% CI 1.06–1.99); (3) VT/kg IBW was not associated with a significant increase of the risk of death (OR 0.92, 95% CI 0.55–1.52). Multivariable analysis confirmed these findings. Conclusions: Crs < 48 ml/cmH2O was associated with ICU mortality, while DP was linearly associated with mortality. DP should be kept as low as possible, even in the case of relatively preserved Crs, irrespective of VT/kg IBW, to reduce the risk of death.

Multimedia archives face the problem of obsolescing and degrading analogue media (e.g., speech and music recordings and video art). In response, researchers in the field have recently begun studying ad hoc tools for the preservation and access of historical analogue documents. This paper investigates the active preservation process of audio tape recordings, specifically focusing on possible means for compensating equalization errors introduced in the digitization process. If the accuracy of corrective equalization filters is validated, an archivist or musicologist would be able to experience the audio as a historically authentic document such that their listening experience would not require the recovery of the original analogue audio document or the redigitization of the audio. Thus, we conducted a MUSHRA-inspired perception test (n = 14) containing 6 excerpts of electronic music (3 stimuli recorded NAB and 3 recorded CCIR). Participants listened to 6 different equalization filters for each stimulus and rated them in terms of similarity. Filters included a correctly digitized “Reference,”an intentionally incorrect “Foil”filter, and a subsequent digital correction of the Foil filter that was produced with a MATLAB script. When stimuli were collapsed according to their filter type (NAB or CCIR), no significant differences were observed between the Reference and MATLAB correction filters. As such, the digital correction appears to be a promising method for compensation of equalization errors although future study is recommended, specifically containing an increased sample size and additional correction filters for comparison.

Nowadays, computer vision is bringing benefits in different scenarios, such as home robotics, autonomous driving and healthcare. The latter is the main application scenario of this work. In this paper, we propose a simplified implementation of a state-of-the-art 3D semantic segmentation deep convolutional network used for automatizing the synthesis of orthopedic casts starting from 3D scans of patients’ arms. The proposed network, based on the PointNet deep learning architecture, is capable of recognising and discriminating among several regions of interest on the scan of the patient’s arm, like the regions around the thumb, the wrist and the elbow. Based on such segmented regions it is then possible to extract important measurements and features to synthesize a custom 3D printed cast. The aforementioned task is very specific and difficult to address with standard 3D segmentation algorithms, moreover it requires very specialized human intervention in data collection and preparation. Until now, semantic regions in human body parts are typically manually annotated by experts to ensure the required accuracy. Unfortunately, this process is time-consuming and, possibly, it may limit the amount of data available for data-driven approaches. In this work, we also investigate the usage of data augmentation to deal with such limited datasets and analyze the model performance by means of cross-validation, which highlights how the proposed architecture model can successfully, and with high accuracy, predict the regions of interest. This is an inspiring result for further research on deep models’ adaptation to challenging applications, for which often clear and consistent data collections are not immediately available. Thus, an empirical approach based on pruning the network parameters and layers, together with a consistent data augmentation technique, could be really effective and prove to be the winning approach.

Gilthead sea bream (GSB) (Sparus aurata) have been generally considered to be resistant to viral encephalo-retinopathy. However, the recent increase of the number of outbreaks in sea bream hatcheries caused by nervous necrosis virus (NNV) called RGNNV/SJNNV, a reassortant virus, has encouraged researchers to investigate the disease pathogenesis in this species, with particular emphasis on fish age. For this purpose, experimental challenges of juveniles and larvae were carried out. Four experimental trials were performed by infecting GSB of different ages with the RGNNV/SJNNV virus: juveniles of 7 g of weight and larvae of 70, 35 and 21 days post hatch (dph). Virological, histological and immunohistochemical analysis, as well as qualitative and quantitative real time PCR at different time points post infection (pi) were performed. ELISA test for antibody detection was also implemented, where applicable. Overall results showed that the gilthead seabream larvae are susceptible to RGNNV/SJNNV infection irrespective to the age, as proved by the massive viral replication detected by quantitative RT-PCR and the massive presence of immunoprecipitates in the nervous tissues evidenced by immunohistochemistry. However, clear clinical signs and mortality were observed only in the youngest group of 21-dph larvae. The viral kinetics investigations led to think that the efficiency of RGNNV/SJNNV replication was higher in 21dph larvae compared to the older ones, which in turn correlated with the disease outcome. Most likely the latter was related to the developmental stage of the immune system. Noteworthy survivors were always tested negative for the presence of specific antibodies. The identification of the factors governing the host ability to control viral replication, or conversely those inhibiting the virus in older fish will be crucial to better understand the disease pathogenesis in this species. Notably, despite the age and the disease outcome of the infection, GSB remained persistently infected for a long time, even up to one year, becoming asymptomatic carriers.