Search

In-distribution and out-distribution datasets Release encompassing both in-distribution and out-distribution datasets associated with the research paper titled "Stylometry for Real-World Expert Coders: a Zero-shot Approach." The in-distribution dataset follows the same experimental setup as the train, validation, and test sets, whereas the out-distribution dataset exclusively comprises snippets used for testing purposes.

This releasecontains three trained models (checkpoints) related to the "Stylometry for Real-World Expert Coders: a Zero-shot Approach" paper, respectively: - MLAllVocaBSoftAtt referees to the soft attention model trained with infoNCE loss without B.P.E, with bounding. - SoftAtt64k104AuthClass referees to the soft-attention model trained in a classification setup with 64k B.P.E. token, with bounding. - SelfAtt64k104authClass referees to the self-attention model

In this paper, we propose to exploit the side-tuning framework for multimodal document classification. Side-tuning is a methodology for network adaptation recently introduced to solve some of the problems related to previous approaches. Thanks to this technique it is actually possible to overcome model rigidity and catastrophic forgetting of transfer learning by fine-tuning. The proposed solution uses off-the-shelf deep learning architectures leveraging the side-tuning framework to combine a base model with a tandem of two side networks. We show that side-tuning can be successfully employed also when different data sources are considered, e.g. text and images in document classification. The experimental results show that this approach pushes further the limit for document classification accuracy with respect to the state of the art., Comment: 2020 25th International Conference on Pattern Recognition (ICPR)

Accountability for global health issues such as a pandemic and its devastating consequences are usually ascribed to a virus, but a comprehensive view should also take into account the state of the host. Data suggests that excessive nutrition is to blame for a yet unknown but not negligible portion of deaths attributed to severe acute respiratory syndrome coronavirus 2. We analyzed the correlation between mean body mass index (BMI) and 2-year coronavirus disease 2019 (COVID-19) mortality rates reported by 181 countries worldwide. Almost two thirds of the countries included had a mean BMI greater or equal to 25, with death rates ranging from 3 to 6,280 per million. Death rates in countries with a mean BMI below 25 ranged from 3 to 1,533. When the analysis was restricted to countries where the extent of testing was deemed more representative of actual mortality, only 20.1% had a mean BMI

Knowledge Tracing is a well-known problem often tackled through purely data-driven approaches. In recent years, many advances have been made in research with various machine learning and deep learning techniques. Despite their satisfactory performance, they have some pitfalls, e.g. modeling one skill at a time, ignoring the relationships between different skills, or inconsistency for the predictions, i.e. sudden spikes and falls across time steps. For this reason, hybrid machine-learning techniques have also been explored. With this systematic literature review, we aim to illustrate this field's state of the art. Specifically, we want to identify the potential and the frontiers in integrating prior knowledge sources in the traditional machine learning pipeline as a supplement to the normally considered data. We applied a qualitative analysis to distill a taxonomy with three dimensions: knowledge source, knowledge representation, and knowledge integration. Exploiting this taxonomy, we also conducted a quantitative analysis to detect the most common approaches.

Background and study aims Endoscopic ultrasound-guided through-the-needle biopsy (TTNB) of pancreatic cystic lesions (PCLs) is associated with a non-negligible risk for adverse events (AEs). We aimed to identify the hierarchic interaction among independent predictors for TTNB-related AEs and to generate a prognostic model using recursive partitioning analysis (RPA). Patients and methods Multicenter retrospective analysis of 506 patients with PCLs who underwent TTNB. RPA of predictors for AEs was performed and the model was validated by means of bootstrap resampling. Results Mean cysts size was 36.7 mm. Most common diagnoses were intraductal papillary mucinous neoplasm (IPMN, 45 %), serous cystadenoma (18.8 %), and mucinous cystadenoma (12.8 %). Fifty-eight (11.5 %) AEs were observed. At multivariate analysis, age (odds ratio [OR] 1.32, 1.09–2.14; p = 0.05), number of TTNB passes (OR from 2.17, 1.32–4.34 to OR 3.16, 2.03–6.34 with the increase of the number of passes), complete aspiration of the cyst (OR 0.56, 0.31–0.95; p = 0.02), and diagnosis of IPMN (OR 4.16, 2.27–7.69; p Conclusion TTNB should be selectively used in the evaluation of patients with IPMN. The present model could be applied during patient selection as to optimize the benefit/risk of TTNB.

To develop and validate a risk score for predicting clinical success after duodenal stenting using self-expanding metallic stents (SEMS) for malignant gastric outlet obstruction (GOO).Consecutive patients who underwent duodenal stenting for malignant GOO were evaluated. Potential predictors of clinical success were determined by uni/multivariate logistic regression analysis.Multiplication of the regression coefficients of the logistic regression model by a factor of two and rounding to obtain easy-to-use point numbers enabling the calculation of the score. Using 10-fold cross-validation, the model was internally validated.One hundred twelve patients were included. Clinical success was achieved in 93 (83.0%) patients. On multivariate logistic regression, selected age ≤65 years (p = 0.05, 1.5 points), stenosis type I (p = 0.04, 3 points), and pancreatic cancer (p = 0.01, 3.5 points) were significant predictors of clinical success. On the Receiver Operating Characteristic (ROC) analysis, a score of 5 had higher specificity and sensitivity.Our score could be useful at identifying, among poor surgical candidates, patients more likely to benefit from SEMS.

Background and Aims Electrocautery-tip lumen-apposing metal stents (EC-LAMSs) have extended the indications of therapeutic EUS. We aimed to retrospectively evaluate safety and technical and clinical success of a newly developed EC-LAMS, the Hot-Spaxus (Taewoong Medical Co, Gimpo, Korea), for various EUS-guided procedures. Methods We included and retrospectively analyzed consecutive patients at 8 tertiary care referral centers who had undergone EUS interventional procedures using the Hot-Spaxus between October 2018 and February 2021. Results Of 58 included patients (male-to-female, 36:22; mean age, 63.5 ± 14.9 years), 29 had undergone pancreatic fluid collection drainage (50%), 22 (37.9%) biliary drainage for malignant distal obstruction, 3 (5.1%) gallbladder drainage for acute cholecystitis, 3 gastroenteroanastomoses, and 1 (1.7%) pelvic collection drainage. Technical success was achieved in 54 of 58 patients (93.1%) and clinical success in all 58. Adverse events occurred in 6 patients (11.1%): 2 early (3.7%), 1 late (1.8%), and 3 long term (5.6%). The outcomes were similar to those observed in a control group of patients treated with the Hot-Axios (Boston Scientific, Marlborough, Mass, USA), the other available EC-LAMS. Conclusions Our study showed that the novel EC-LAMS has high technical and clinical success rates for various interventional EUS indications. Future multicenter prospective studies will better clarify the role of this new EC-LAMS for different indications.

The recent development of Quantum Cascade Lasers (QCLs) represents one of the biggest opportunities for the deployment of a new class of Free Space Optical (FSO) communication systems working in the mid-infrared (mid-IR) wavelength range. As compared to more common FSO systems exploiting the telecom range, the larger wavelength employed in mid-IR systems delivers exceptional benefits in case of adverse atmospheric conditions, as the reduced scattering rate strongly suppresses detrimental effects on the FSO link length given by the presence of rain, dust, fog, and haze. In this work, we use a novel FSO testbed operating at 4.7 µm, to provide a detailed experimental analysis of noise regimes that could occur in realistic FSO mid-IR systems based on QCLs. Our analysis reveals the existence of two distinct noise regions, corresponding to different realistic channel attenuation conditions, which are precisely controlled in our setup. To relate our results with real outdoor configurations, we combine experimental data with predictions of an atmospheric channel loss model, finding that error-free communication could be attained for effective distances up to 8 km in low visibility conditions of 1 km. Our analysis of noise regimes may have a key relevance for the development of novel, long-range FSO communication systems based on mid-IR QCL sources.

Objective Selpercatinib is a highly selective RET-inhibitor drug, approved for the treatment of RET-altered lung and thyroid cancers. So far, RET-altered medullary thyroid cancer (MTC) patients treated with selpercatinib showed a remarkable objective response rate and safety profile. However, new treatment emerging adverse events (TEAEs) have been recently reported. The aim of this study was to evaluate the prevalence, features, and clinical management of effusions that are one of these TEAEs. Design Around 10 of 11 patients with advanced MTC enrolled in the LIBRETTO-201 clinical trial at Endocrinology Unit of the Pisa University Hospital were evaluated for the presence and management of effusions. Methods We retrospectively evaluated MTC patients treated with selpercatinib. The presence of pleural, pericardial, abdominal, and/or pelvic effusions was evaluated by reviewing the computerized tomography scan performed during the study protocol and up to 24 months of observation. Results All but one MTC patient experienced previous multikinase inhibitors treatment. Three patients already had effusions before starting selpercatinib treatment. New effusions appeared in eight of ten (80%) patients during the treatment. A chylous nature was documented in patients who underwent fluid aspiration. Whenever a dose reduction was performed, a significant positive effect was observed. Conclusions Chylous effusions are a new TEAE of selpercatinib treatment. They can appear or worsen at any time during the treatment. For cases with asymptomatic and mild effusions, active surveillance may be appropriate and safe. In symptomatic and/or moderate/severe cases, aspiration of the fluid and a dose reduction can improve this AE, strongly supporting a cause-effect correlation with selpercatinib. Significance statement Effusions, particularly of chylous nature, represent emergent and quite frequent adverse events in the management of patients affected by advanced MTC on treatment with the highly selective inhibitor selpercatinib. In this study, we evaluated, in a series of MTC patients treated with selpercatinib, the prevalence of pleural, pericardial, abdominal, and/or pelvic effusions. Insights into the diagnosis and treatment of the effusions are provided as well as suggestions for clinical management.

SUNNY is a k-nearest neighbors based Algorithm Selection (AS) approach that schedules and runs a number of solvers for a given unforeseen problem. In this work we present sunny-as2, an enhancement of SUNNY for generic AS scenarios that advances the original approach with wrapper-based feature selection, neighborhood-size configuration and a greedy approach to speed-up the training phase. Empirical evidence shows that sunny-as2 is competitive w.r.t. state-of-the-art AS approaches.

During COVID-19 pandemic, lung ultrasound (LUS) proved to be of great value in the diagnosis and monitoring of patients with pneumonia. However, limited data exist regarding its use to assess aeration changes during follow-up (FU). Our study aims to prospectively evaluate 232 subjects who underwent a 3-month-FU program after hospitalization for COVID-19 at the University Hospital of Pisa. The goals were to assess the usefulness of standardized LUS compared with the gold standard chest computed tomography (CT) to evaluate aeration changes and to verify LUS and CT agreement at FU. Patients underwent in the same day a standardized 16-areas LUS and high-resolution chest CT reported by expert radiologists, assigning interpretative codes. Based on observations distribution, LUS score cut-offs of 3 and 7 were selected, corresponding to the 50th and 75th percentile, respectively. Patients with LUS scores above both these thresholds were older and with longer hospital stay. Patients with a LUS score ≥3 had more comorbidities. LUS and chest CT showed a high agreement in identifying residual pathological findings, using both cut-off scores of 3 (OR 14,7; CL 3,6-64,5, Sensitivity 91%, Specificity 49%) and 7 (OR 5,8; CL 2,3-14,3, Sensitivity 65%, Specificity 79%). Our data suggest that LUS is very sensitive in identifying pathological findings at FU after a hospitalization for COVID-19 pneumonia, compared to CT. Given its low cost and safety, LUS could replace CT in selected cases, such as in contexts with limited resources or it could be used as a gate-keeper examination before more advanced techniques.

Cardiovascular diseases have a high prevalence and incidence and are associated with a high rate of mortality and hospitalization; this is reflected in a reduction in quality of life and a high health care expenditure. The aim of our study was to evaluate the effectiveness of telemedicine in primary and secondary prevention, through the use of ECG, ABPM and Holter monitoring. Health Telematic Network, in collaboration with the University of Brescia and Federfarma, has installed in 6527 territorial pharmacies distributed in the country, a telematic network interconnected to a single Telemedicine platform, where cardiologists are available for H24 reporting and teleconsultation. The pharmacies performed 12–lead ECG in healthy subjects, with cardiovascular risk factors or ongoing symptoms referable to the cardiovascular system, ABPM in accordance with the guidelines for hypertension and Holter ECG in symptomatic patients (heart palpitations, syncope, lipothymias) or positive history for brady and tachyarrhythmias. In the period between 2016 and October 2021, we analyzed 660508 exams: 411526 ECG, 143869 cardiac Holter and 105113 ABPM. Of the 411526 ECGs, 14.5% of tracings showed abnormalities not compatible with the reported anamnestic data, therefore 59671 users were sent to the caregiver for further evaluation or diagnostic investigation; in 1.5% of cases, major abnormalities were detected, necessitating immediate referral of the patient to the emergency department. Of the 105113 ABPM, 37.2% of the examinations revealed abnormal blood pressure trends according to the SIIA classification, 39102 users were sent to the physician for the appropriate diagnostic and therapeutic measures. Of the 143869 Holters, 26326 patients (18.3%) were referred to the attending physician for major arrhythmias, 3008 patients (2.1%) were referred to the emergency department for red code arrhythmias (VT, grade 3 BAV, long QTc, high or low ventricular response AF with critical pauses). Our data show the important role of a Telemedicine network in primary and secondary cardiovascular prevention, able to provide early diagnosis and treatment of major events; the use of a national telemedicine network has also proved effective in the appropriate referral of subjects to the Emergency Department, with a positive impact in terms of health care expenditure containment.

During the Italian national lockdown (11/3 to 4/5/2020), the COVID–19 epidemic caused a shift of hospital resources towards the management of the COVID+ patient; hospitals saw a progressive drop of admissions for acute myocardial infarction and arrhythmias. In order to verify the role of Telemedicine in this context, we compared 6104 ECGs recorded by 5000 pharmacies, with the 17280 ECGs of the same period of 2019, evaluated and stored in the telemedicine platform of Health Telematic Network (HTN), in collaboration with the Department of Cardiology of the University of Brescia, Federfarma and Istituto Superiore di Sanità. Chest pain was the reason for examination in 298 patients (4.88%) vs 402 (2.33%) in 2019 (+109.86%). Of these, 118 were in Lombardy, (39.50% of the total), 36 in the province of Brescia (30.50%) and 28 in the province of Bergamo (23.73%). 8 ECGs showed a STEMI picture, (2.68%), compared with 7 STEMIs (1.74%) detected in 2019 (+ 54.17%). All of these patients were referred urgently to the emergency department. We detected Atrial Fibrillation in 344 patients (5.64%), vs 393 (2.27%) in 2019, (+ 40.25%); 32 cases of Atrial Flutter (0.52%), vs 25 (0.14%) detected in 2019 (+ 26.92%). Finally, we found paroxysmal supraventricular tachycardia in 8 cases (0.13%), vs 6 cases (0.03%) detected in 2019(+ 23.07%). During the lockdown period, 384 patients (6.29%) were referred to the emergency department for symptomatic tachyarrhythmia, vs 424 patients (3.47%) in 2019(+ 55.16%). In Lombardy, 194 cases of tachyarrhythmia were reported in territorial pharmacies (about 50.52% of the total in Italy), 93 in the Brescia area (47.94%), 50 in the Bergamo area (25.77%). Conclusions: a high number of patients with cardiovascular symptoms preferred to refer to territorial pharmacies during the COVID–19 blast. Telemedicine allowed patients with STEMI to access hospitals quickly, avoiding a serious diagnostic delay; a significant proportion of accesses to local pharmacies for chest pain occurred in the region most affected by the COVID–19 outbreak; the service also allowed to refer to the hospital only patients with clinically relevant tachyarrhythmias, avoiding delay of care and inappropriate accesses. This underlines how the telemedicine network provided by pharmacies can become an important resoruce for the citizen, supporting the services of the Italian National Health System.

Studies comparing EUS-guided fine-needle aspiration (EUS-FNA) with EUS-guided fine-needle biopsy (EUS-FNB) for the evaluation of pancreatic neuroendocrine tumors (pNETs) are lacking. We aimed at comparing EUS-FNA with EUS-FNB in terms of Ki-67 proliferative index (PI) estimation capability, cellularity of the samples, and reliability of Ki-67 PI/tumor grading compared with surgical specimens.Patients diagnosed with pNETs on EUS and/or surgical specimens were retrospectively identified. Specimens were re-evaluated to assess Ki-67 PI feasibility, sample cellularity by manual counting, and determination of Ki-67 PI value. Outcomes in the EUS-FNA and EUS-FNB groups were compared. Kendall rank test was used for Ki-67 PI correlation between EUS and surgical specimens. Subgroup analysis including small (≤20 mm), non-functioning pNETs was performed.Three-hundred samples from 292 lesions were evaluated: 69 EUS-FNA cytology and 231 EUS-FNB histology. Ki-67 PI feasibility was similar for EUS-FNA and EUS-FNB (91.3% vs. 95.7%, p = 0.15), while EUS-FNB performed significantly better in the subgroup of 179 small pNETs (88.2% vs. 96.1%, p = 0.04). Rate of poor cellulated (500 cells) specimens was equal between EUS-FNA and EUS-FNB. A significant correlation for Ki-67 PI values between EUS and 92 correspondent surgical specimens was found in both groups, but it was stronger with EUS-FNB (tau = 0.626, p 0.0001 vs. tau = 0.452, p = 0.031). Correct grading estimation was comparable between the two groups (p = 0.482).Our study showed stronger correlation for Ki-67 values between EUS-FNB and surgical specimens, and that EUS-FNB outperformed EUS-FNA in the evaluation of small pNETs. EUS-FNB should become standard of care for grading assessment of suspected pNETs.

In this article, we propose a new robust, stable, accurate, and efficient time-delay estimation scheme for the angle-of-arrival (AoA) ultrasonic indoor single-source localization systems. The AoA technique employing microphone arrays and ultrasonic signal emitting sources has turned out to be a promising alternative to the conventional distance-based ultrawideband systems for indoor localization. The crux of AoA estimation lies in the derivation of time-delay estimates, which is typically handled by advanced digital signal processing approaches based on cross-correlation and matching algorithms. The calculation of the channel response for each received signal and the conversion from the time to the time-delay domain consent our echo suppression delay estimator (ESDE) to apply echo suppression, in order to overcome the typical challenges introduced by the channel in indoor environments. Moreover, the optimization of the cross-correlation algorithm through oversampling the cross power spectrum and the curve-fitting algorithm allows to ensure subsample precision results. The performances of our developed AoA estimation system are evaluated by carrying out real-world experiments and compared with the state-of-the-art generalized cross correlators. Our ESDE algorithm is shown to outperform several generalized solutions.

Quantum cascade laser frequency combs are nowadays well-appreciated sources for infrared spectroscopy. Here their applicability for free-space optical communication is demonstrated. The spontaneously-generated intermodal beat note of the frequency comb is used as carrier for transferring the analog signal via frequency modulation. Exploiting the atmospheric transparency window at 4 $\mu$m, an optical communication with a signal-to-noise ratio up to 65 dB is realized, with a modulation bandwidth of 300 kHz. The system tolerates a maximum optical attenuation exceeding 35 dB. The possibility of parallel transmission of an independent digital signal via amplitude modulation at 5 MS/s is also demonstrated., Comment: 17 pages, 13 figures

International audience; The paper presents the SEAWALL platform enabling heterogeneous data acquisition and low-latency processing for the Industry 4.0, developed within the homonymous project founded by the Italian BIREX industrial consortium. The architecture features cutting-edge technologies (such as Kubernetes, ISTIO, KubeEdge, W3C WoT), in order to support the seamless orchestration of workloads among the nodes of a cloud-edge continuum in QoS-aware scenarios where the latency requirement of the anomaly detection must be continuously assessed. The paper presents the industrial use-case from the SEAWALL project and the components of the cloud/edge architecture.

Introduction: The dynamism in the regulatory frameworks concerning the consumption of cannabinoids has placed their effects on cognitive and psychomotor skills at the center of the scientific debate. In consideration of the potential repercussions on public safety, particular attention has been focused on the impairment of driving skills, opening up the need to specifically regulate driving under the effects of cannabinoids. Pharmacokinetics: Both native cannabinoids and metabolites show a long positivity at low concentrations in the biological fluids, especially in the case of chronic consumption. Qualitative positivity to cannabinoids does not itself prove the presence of detrimental effects, which require the presence of active substances at relevant concentrations. Driving Skill Impairment: Multiple studies highlight a tetrahydrocannabinol (THC) concentration- based alteration of driving skills mainly affecting automatisms, whereas skills subjected to cognitive control are preserved up to higher dosages. The evidence relating to associations with other substances, chronic consumption and other cannabinoids, on the other hand, is still burdened by a high degree of uncertainty. Regulation Policies: Different models can be adopted in the regulation of driving under the effects of cannabinoids: sanctions can be applied in case of qualitative positivity, cannabinoids concentration above a defined threshold, or in presence of a demonstrated state of cognitive alteration. Conclusion: "Per se limit" with a quantitative THC cut-off between 3.5 and 5 ng/ml can currently be considered the most balanced choice. Finally, the analysis carried out allowed to identify pitfalls in both scientific and legislative fields for the improvement of safety policies

OBJECTIVES Type 1 autoimmune pancreatitis (AIP) is a manifestation of immunoglobulin G4-related diseases (IgG4-RD). To evaluate the activity of the disease, the IgG4-RD responder index (RI) has been created. This study evaluated the IgG4-RD RI as prognostic factor of 1-year disease relapse. METHODS Patients diagnosed with type 1 AIP between January 2012 and December 2016, with available magnetic resonance imaging and IgG4 dosage, were enrolled. Immunoglobulin G4-RD RI was calculated at baseline (time 0), and at 3 to 6 and 12 to 18 months after the end of steroid therapy (time 1 and time 2, respectively). RESULTS Thirty-three patients were included in the study. Immunoglobulin G4-RD RI was 8.9 (standard deviation [SD], 3.8) at time 0, 2.4 (SD, 3.1) at time 1 (P < 0.0001 vs time 0), and 4.2 (SD, 3.9) at time 2 (P = 0.02 vs time 1). Fourteen patients who relapsed within 1 year showed a higher mean value of IgG4-RD RI at time 0 (10.9; SD, 4.3) versus 19 who did not (7.4; SD, 2.6; P = 0.012). This difference was observed also at time 2 (6.8 vs 2.1; P = 0.002). CONCLUSIONS Immunoglobulin G4-RD RI correlates with type 1 AIP disease activity, and it predicts disease relapse within 1 year.

Movements, habitat use, and activity of herbivorous species are mainly influenced by trade-offs between food availability and daily/seasonal variation in predation risk. Nevertheless, studies conducted across several seasons and evaluating both spatial and temporal responses of meso-small herbivores are still scanty. Additionally, information on spatiotemporal behaviour is often lacking for cryptic, localised species, which may limit conservation actions. We evaluated how sex, time of day, seasonality, and habitat type—reflecting trade-offs in predation risk and access to food resources—influenced the spatial behaviour and locomotor activity of an herbivorous mammal, the Apennine hare, a threatened species endemic to central-southern Italy. In a 4-year study, we intensively radio-tracked 12 individuals, providing the first insights on the spatiotemporal ecology of this lagomorph. Sex affected neither home range size nor habitat selection. Home range size was larger during the night and in warm months, when hares moved to the most energetically rewarding habitat type for feeding activities, i.e., cultivations. At both study area and home range spatial scales, habitat selection did not vary between the cold and the warm months. At the study area scale, hares avoided deciduous woodland and human settlements, whereas selected the Mediterranean scrubwood and cultivations. Within home ranges, at night, Apennine hares selected risky patches, i.e., cultivations as feeding grounds, whereas in daylight, the safer Mediterranean scrubwood was used for cover and resting. Throughout the year, Apennine hares were mostly nocturnal and were more active in open than in concealed habitats, with no differences between sexes. Bright nights, i.e., with full moon and clear sky, inhibited activity in open areas but not in cover, likely to reduce predation risk. Our findings emphasised the role of feeding/antipredatory requirements in shaping spatiotemporal behaviour of meso-small mammals, with potential consequences for the conservation of threatened species.

Background: Long COVID has become a burden on healthcare systems worldwide. Research into the etiology and risk factors has been impeded by observing all diverse manifestations as part of a single entity. We aimed to determine patterns of symptoms in convalescing COVID-19 patients. Methods: Symptomatic patients were recruited from four countries. Data were collected regarding demographics, comorbidities, acute disease and persistent symptoms. Factor analysis was performed to elucidate symptom patterns. Associations of the patterns with patients’ characteristics, features of acute disease and effect on daily life were sought. Results: We included 1027 symptomatic post-COVID individuals in the analysis. The majority of participants were graded as having a non-severe acute COVID-19 (N = 763, 74.3%). We identified six patterns of symptoms: cognitive, pain-syndrome, pulmonary, cardiac, anosmia-dysgeusia and headache. The cognitive pattern was the major symptoms pattern, explaining 26.2% of the variance; the other patterns each explained 6.5–9.5% of the variance. The cognitive pattern was higher in patients who were outpatients during the acute disease. The pain-syndrome pattern was associated with acute disease severity, higher in women and increased with age. The pulmonary pattern was associated with prior lung disease and severe acute disease. Only two of the patterns (cognitive and cardiac) were associated with failure to return to pre-COVID occupational and physical activity status. Conclusion: Long COVID diverse symptoms can be grouped into six unique patterns. Using these patterns in future research may improve our understanding of pathophysiology and risk factors of persistent COVID, provide homogenous terminology for clinical research, and direct therapeutic interventions.

Meta-solver approaches exploit many individual solvers to potentially build a better solver. To assess the performance of meta-solvers, one can adopt the metrics typically used for individual solvers (e.g., runtime or solution quality) or employ more specific evaluation metrics (e.g., by measuring how close the meta-solver gets to its virtual best performance). In this paper, based on some recently published works, we provide an overview of different performance metrics for evaluating (meta-)solvers by exposing their strengths and weaknesses.

Background and study aims Endoscopic ultrasound-guided through-the-needle biopsy (TTNB) of pancreatic cystic lesions (PCLs) is associated with a non-negligible risk for adverse events (AEs). We aimed to identify the hierarchic interaction among independent predictors for TTNB-related AEs and to generate a prognostic model using recursive partitioning analysis (RPA). Patients and methods Multicenter retrospective analysis of 506 patients with PCLs who underwent TTNB. RPA of predictors for AEs was performed and the model was validated by means of bootstrap resampling. Results Mean cysts size was 36.7mm. Most common diagnoses were intraductal papillary mucinous neoplasm (IPMN, 45%), serous cystadenoma (18.8%), and mucinous cystadenoma (12.8%). Fifty-eight (11.5%) AEs were observed. At multivariate analysis, age (odds ratio [OR] 1.32, 1.09-2.14; p=0.05), number of TTNB passes (OR from 2.17, 1.32-4.34 to OR 3.16, 2.03-6.34 with the increase of the number of passes), complete aspiration of the cyst (OR 0.56, 0.31-0.95; p=0.02), and diagnosis of IPMN (OR 4.16, 2.27-7.69; p

We demonstrate the presence of intensity correlation in the harmonic comb emitted by quantum cascade lasers, paving the way for a deeper understanding of the role of the four-wave-mixing process enabling this emission.

We demonstrate the potentiality of quantum-cascade-laser frequency combs for optical communication in the atmospheric transparency window around 4 µm. The intermodal beat note is used as carrier for transmitting the analog signal via frequency-modulation.

Analog FM free-space optical communication based on a mid-infrared quantum cascade laser frequency comb: supplemental document

Programming language detection is a common need in the analysis of large source code bases. It is supported by a number of existing tools that rely on several features, and most notably file extensions, to determine file types. We consider the problem of accurately detecting the type of files commonly found in software code bases, based solely on textual file content. Doing so is helpful to classify source code that lack file extensions (e.g., code snippets posted on the Web or executable scripts), to avoid misclassifying source code that has been recorded with wrong or uncommon file extensions, and also shed some light on the intrinsic recognizability of source code files. We propose a simple model that (a) use a language-agnostic word tokenizer for textual files, (b) group tokens in 1-/2-grams, (c) build feature vectors based on N-gram frequencies, and (d) use a simple fully connected neural network as classifier. As training set we use textual files extracted from GitHub repositories with at least 1000 stars, using existing file extensions as ground truth. Despite its simplicity the proposed model reaches ≈ 85% in our experiments for a relatively high number of recognized classes (more than 130 file types).

This paper proposes two efficient and easy-to-use error mitigation solutions to the problem of three-dimensional (3-D) angle-of-arrival (AOA) source localization in the mixed line-of-sight (LOS) and non-line-of-sight (NLOS) indoor environments. A weighted linear least squares estimator is derived first for the LOS AOA components in terms of the direction vectors of arrival, albeit in a sub-optimal manner. Next, data selection exploiting the sum of squared residuals is carried out to discard the error-prone NLOS connections. In so doing, the first approach is constituted and more accurate closed-form location estimates can be obtained. The second method applies a simulated annealing stochastic framework to realize the robust $\ell_1$-minimization criterion, which therefore falls into the methodology of statistical robustification. Computer simulations and ultrasonic onsite experiments are conducted to evaluate the performance of the two proposed methods, demonstrating their outstanding positioning results in the respective scenarios., This paper has been accepted for oral presentation at 2021 International Conference on Indoor Positioning and Indoor Navigation (IPIN), 29 Nov. -- 2 Dec. 2021, Lloret de Mar, Spain

Biomass gasification is a thermochemical process in which the biomass is converted into a mixture of gases, called syngas, commonly utilised in thermal machines to produce electricity and heat. In the present research activity, the conventional air-gasification in downdraft gasifier is replaced by oxy-CO2 gasification technology. This strategy allows to obtain a nitrogen-free syngas, mainly composed by CO, H2, CO2 and CH4, which can be used into the synthesis of various bio-fuels like methanol or synthetic natural gas (SNG). Carbon dioxide is utilized as gasifying agent together with oxygen to mitigate the reactivity of the latter, which can lead to excessive reaction temperatures. In particular, the present work shows the preliminary results of an experimental campaign carried on utilizing a small scale downdraft gasifier (max thermal power of roughly 100 kW) fed with wood pellet and using a mixture of oxygen and CO2 as gasification agent. The experimental results have been utilized to calibrate a pseudo-kinetic model of the oxy-CO2 gasification process, implemented in Aspen Plus environment. To analyse the possibility to transform the produced syngas into methane, an additional numerical model of a methanation plant was then utilised. As a whole, the numerical analysis confirms its usefulness in predicting the performance of the gasifier, which can reach a cold gas efficiency of around 70%, while the methanation plant can achieve a production of roughly 20 kg of methane per 100 kg of gasified wood.