Search

In recent times, with the exception of sporadic cases, the trend in computer vision is to achieve minor improvements compared to considerable increases in complexity. To reverse this trend, we propose a novel method to boost image classification performances without increasing complexity. To this end, we revisited ensembling, a powerful approach, often not used properly due to its more complex nature and the training time, so as to make it feasible through a specific design choice. First, we trained two EfficientNet‐b0 end‐to‐end models (known to be the architecture with the best overall accuracy/complexity trade‐off for image classification) on disjoint subsets of data (i.e., bagging). Then, we made an efficient adaptive ensemble by performing fine‐tuning of a trainable combination layer. In this way, we were able to outperform the state‐of‐the‐art by an average of 0.5% on the accuracy, with restrained complexity both in terms of the number of parameters (by 5–60 times), and the FLoating point Operations Per Second FLOPS by 10–100 times on several major benchmark datasets. [ABSTRACT FROM AUTHOR]

Background Few case reports describe the clinical features of pancreatic involvement in inflammatory bowel disease. Aim To investigate prevalence and disease course of inflammatory bowel disease children with pancreatitis and with exclusive hyperamylasemia and hyperlipasemia. Methods We used a web-registry to retrospectively identify paediatric inflammatory bowel disease patients with hyperamylasemia and hyperlipasemia. Participants were re-evaluated at 6 months and 1 year. Results From a total of 649 paediatric patients, we found 27 with hyperamylasemia and hyperlipasemia (4.1%). Eleven patients (1.6%) fulfilled diagnostic criteria for acute pancreatitis. Female gender was significantly associated with acute pancreatitis ( p = 0.04). Twenty-five children (92.5%) had colonic disease. At 6 months 1/11 children with acute pancreatitis (9%) showed acute recurrent pancreatitis, while 1 patient (9%) had persistent hyperamylasemia and hyperlipasemia. At 12 months, 1 patient showed chronic pancreatitis (9.1%). Of the 16 children with exclusive hyperamylasemia and hyperlipasemia, 4 developed acute pancreatitis (25%), while 1 patient (6.2%) still presented exclusive hyperamylasemia and hyperlipasemia at 6 months. At 12 months, 11/16 patients (68.7%) reached a remission of pancreatic involvement, whereas 5 remaining patients (32.3%) had persistent hyperamylasemia and hyperlipasemia. Conclusions In inflammatory bowel disease children, acute pancreatitis is more common in colonic disease and in female gender. Pancreatic function should be monitored, considering that pancreatic damage may evolve. [ABSTRACT FROM AUTHOR]

This paper discusses the problem of fostering lifestyle changes towards healthier habits via tailored user guidance. We present a novel multisensory device, the Wize Mirror, which will be able to detect semeiotic face signs related to cardio-metabolic risk, and encourage users to reduce their risk by improving their lifestyle. Offering a proper user guidance requires solving three main issues: user profiling, definition of a wellness index based on biophysical data, and personalized guidance by means of coaching and supportive messages. For each of these issues, the solutions proposed in the EU FP7 Project SEMEOTICONS are presented, highlighting their advantages with respect to the state-of-the-art. [ABSTRACT FROM AUTHOR]

To correlate, RSI with SVHI, in a group of singing students, by means of a specific anamnestic questionnaire which analyzes the physical, social, emotional, and economic impacts of voice problems on their lives. This study is a cross-sectional single-center observational study. Forty-two modern singing students (26F/16M; average age: 24.9 ± 5.7; range: 16–46 years old) were recruited. A self-assessment of the singing-voice (SVHI) and of reflux symptom (RSI) was performed. Using the validated RSI threshold, 31% of participants were classified as RSI greater than 13 as an indication of suspected LPR. Classifying the SVHI score as proposed in the recent literature (cutoff of 20.35) our sample was a voice disorders prevalence of 71.4%. There was no significant positive or negative relationship between RSI and SVHI total score (rho = 0.238, P = 0.13). Instead, by considering the relationship between the individual items of the two questionnaires, it is highlighted that the item 1 of SVHI was significantly correlated with most of the RSI items (P = 0.0001–0.006). In the same vein, the item 5 of SVHI was correlated to hoarseness and coughing (P = 0.005–0.006). The item 20 of SVHI was significantly correlated with hoarseness, excess mucus in the throat or postnasal drip and with the pharyngeal Globus sensation (P = 0.001–0.005). By aggregating the SVHI item response as a positive response (2–4) versus a negative response (0 or 1) between "RSI pathology classification," a significant association was observed for SVHI item 1 (P < 0.021), item 5 (P < 0.006), item 20 (P < 0.042), item 24 (P < 0.044) item 25 (P < 0.047). These associations were confirmed by univariate binary logistic. Multivariate binary logistic regression confirms that SVHI Item 1, Item 5 were more associated with RSI positive. Based on results, we propose a questionnaire that combines the most relevant SVHI items correlated to LPRD (SVHI-10-LPRD questionnaire). Cronbach's alpha coefficient for the 10 items selected was 0.87; item-total correlation coefficients for each item were in the range of 0.461 to 0.670. This pilot study shows that, in case of significant RSI for LPR, it is possible to observe a significant association with some symptoms described in SVHI. These results underline that the association of the RSI and SVHI questionnaires administered to singers and singing students, can represent a simple screening to reveal possible alterations of the singing voice correlate to LPR. Moreover, we propose a Singing Voice Handicap Index correlated to LPR (SVHI-10-LPRD). It will be necessary to increase the sample of subjects in the study to confirm these preliminary data. [ABSTRACT FROM AUTHOR]

Copyright of SCIRES-IT is the property of CASPUR-CIBER Publishing and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

The confinement properties of the open structure formed crossing a circular waveguide perpendicular to a parallel-plate waveguide are discussed, highlighting the fundamental differences with respect to the common high-frequency resonators. The structure supports a TE011 mode that appears suited for millimeter- and submillimeter-wave applications. The experimental characterization of this mode in a configuration resonating at 281 GHz is described in detail. The observed resonance exhibits state-of-the-art performances. The properties of the TE011 mode are studied in terms of the geometry, calculating the mode chart and the related quality factor and power-to-field conversion efficiency. The mode chart is then determined for configurations including a sample holder, where one of the waveguides is filled with a low-loss dielectric material. In addition to excellent merit figures, the TE011 mode shows a relevant stability with respect to the geometrical imperfections and to the insertion of a sample holder. The obtained results demonstrate that the proposed single-mode resonator competes with the standard cavities in terms of performances, versatility, and simplicity. [ABSTRACT FROM AUTHOR]

Exploiting hand gestures for non-verbal communication has extraordinary potential in HCI. A data glove is an apparatus widely used to recognize hand gestures. To improve the functionality of the data glove, a highly stretchable and reliable signal-to-noise ratio sensor is indispensable. To do this, the study focused on the development of soft silicone microchannel sensors using a Eutectic Gallium-Indium (EGaIn) liquid metal alloy and a hand gesture recognition system via the proposed data glove using the soft sensor. The EGaIn-silicone sensor was uniquely designed to include two sensing channels to monitor the finger joint movements and to facilitate the EGaIn alloy injection into the meander-type microchannels. We recruited 15 participants to collect hand gesture dataset investigating 12 static hand gestures. The dataset was exploited to estimate the performance of the proposed data glove in hand gesture recognition. Additionally, six traditional classification algorithms were studied. From the results, a random forest shows the highest classification accuracy of 97.3% and a linear discriminant analysis shows the lowest accuracy of 87.4%. The non-linearity of the proposed sensor deteriorated the accuracy of LDA, however, the other classifiers adequately overcame it and performed high accuracies (>90%). [ABSTRACT FROM AUTHOR]

An amendment to this paper has been published and can be accessed via the original article. [ABSTRACT FROM AUTHOR]