Search

Your search keyword '"Odone, A."' showing total 4,975 results
Start Over Author "Odone, A."
4,975 results on '"Odone, A."'

1. Transferring disentangled representations: bridging the gap between synthetic and real images

Author

Dapueto, Jacopo, Noceti, Nicoletta, and Odone, Francesca

Subjects
Computer Science - Computer Vision and Pattern Recognition, Computer Science - Artificial Intelligence
Abstract

Developing meaningful and efficient representations that separate the fundamental structure of the data generation mechanism is crucial in representation learning. However, Disentangled Representation Learning has not fully shown its potential on real images, because of correlated generative factors, their resolution and limited access to ground truth labels. Specifically on the latter, we investigate the possibility of leveraging synthetic data to learn general-purpose disentangled representations applicable to real data, discussing the effect of fine-tuning and what properties of disentanglement are preserved after the transfer. We provide an extensive empirical study to address these issues. In addition, we propose a new interpretable intervention-based metric, to measure the quality of factors encoding in the representation. Our results indicate that some level of disentanglement, transferring a representation from synthetic to real data, is possible and effective.

Published
2024

2. Self-supervised pre-training with diffusion model for few-shot landmark detection in x-ray images

Author

Di Via, Roberto, Odone, Francesca, and Pastore, Vito Paolo

Subjects
Computer Science - Computer Vision and Pattern Recognition, Computer Science - Artificial Intelligence
Abstract

In the last few years, deep neural networks have been extensively applied in the medical domain for different tasks, ranging from image classification and segmentation to landmark detection. However, the application of these technologies in the medical domain is often hindered by data scarcity, both in terms of available annotations and images. This study introduces a new self-supervised pre-training protocol based on diffusion models for landmark detection in x-ray images. Our results show that the proposed self-supervised framework can provide accurate landmark detection with a minimal number of available annotated training images (up to 50), outperforming ImageNet supervised pre-training and state-of-the-art self-supervised pre-trainings for three popular x-ray benchmark datasets. To our knowledge, this is the first exploration of diffusion models for self-supervised learning in landmark detection, which may offer a valuable pre-training approach in few-shot regimes, for mitigating data scarcity.

Published
2024

3. Looking at Model Debiasing through the Lens of Anomaly Detection

Author

Pastore, Vito Paolo, Ciranni, Massimiliano, Marinelli, Davide, Odone, Francesca, and Murino, Vittorio

Subjects
Computer Science - Machine Learning, Computer Science - Computer Vision and Pattern Recognition, I.4, I.5
Abstract

It is widely recognized that deep neural networks are sensitive to bias in the data. This means that during training these models are likely to learn spurious correlations between data and labels, resulting in limited generalization abilities and low performance. In this context, model debiasing approaches can be devised aiming at reducing the model's dependency on such unwanted correlations, either leveraging the knowledge of bias information or not. In this work, we focus on the latter and more realistic scenario, showing the importance of accurately predicting the bias-conflicting and bias-aligned samples to obtain compelling performance in bias mitigation. On this ground, we propose to conceive the problem of model bias from an out-of-distribution perspective, introducing a new bias identification method based on anomaly detection. We claim that when data is mostly biased, bias-conflicting samples can be regarded as outliers with respect to the bias-aligned distribution in the feature space of a biased model, thus allowing for precisely detecting them with an anomaly detection method. Coupling the proposed bias identification approach with bias-conflicting data upsampling and augmentation in a two-step strategy, we reach state-of-the-art performance on synthetic and real benchmark datasets. Ultimately, our proposed approach shows that the data bias issue does not necessarily require complex debiasing methods, given that an accurate bias identification procedure is defined., Comment: 15 pages, 7 figures

Published
2024

4. Is in-domain data beneficial in transfer learning for landmarks detection in x-ray images?

Author

Di Via, Roberto, Santacesaria, Matteo, Odone, Francesca, and Pastore, Vito Paolo

Subjects
Computer Science - Computer Vision and Pattern Recognition
Abstract

In recent years, deep learning has emerged as a promising technique for medical image analysis. However, this application domain is likely to suffer from a limited availability of large public datasets and annotations. A common solution to these challenges in deep learning is the usage of a transfer learning framework, typically with a fine-tuning protocol, where a large-scale source dataset is used to pre-train a model, further fine-tuned on the target dataset. In this paper, we present a systematic study analyzing whether the usage of small-scale in-domain x-ray image datasets may provide any improvement for landmark detection over models pre-trained on large natural image datasets only. We focus on the multi-landmark localization task for three datasets, including chest, head, and hand x-ray images. Our results show that using in-domain source datasets brings marginal or no benefit with respect to an ImageNet out-of-domain pre-training. Our findings can provide an indication for the development of robust landmark detection systems in medical images when no large annotated dataset is available., Comment: Accepted to ISBI 2024

Published
2024

9. Impact of PRECEDE–PROCEED Model Audits in Cancer Screening Programs in Lombardy Region: Supporting Equity and Quality Improvement

Author

Stefano Odelli, Margherita Zeduri, Maria Rosa Schivardi, Davide Archi, Liliana Coppola, Roberto Genco Russo, Maristella Moscheni, Elena Tettamanzi, Fabio Terragni, Michela Viscardi, Valentina Vitale, Anna Odone, Danilo Cereda, and Silvia Deandrea

Subjects
PRECEDE–PROCEED model, cancer screening, audit, Lombardy region, equity, quality improvement, Neoplasms. Tumors. Oncology. Including cancer and carcinogens, RC254-282
Abstract

Background: Health disparities related to socio-economic factors impact access to preventive health interventions. The PRECEDE–PROCEED model, a multidimensional approach to health promotion, has been adapted to optimise cancer screening programs in Lombardy, Italy, addressing these disparities. Methods: This study evaluated the application of systemic audits based on the PRECEDE–PROCEED model across Lombardy cancer screening programs. A systematic region-wide audit was performed in 2019, and follow-up audits were performed in 2022–2023. Data were collected using structured analysis methodologies, including epidemiological, behavioural, and organisational assessments. Results: The 2019 audit showed strengths in participation and quality standards but identified challenges in cervical cancer screening coverage and waiting times for assessments. Improvements plans included the digitisation of processes and stakeholder engagement. The 2022–2023 audits reported increased coverage for breast and colorectal screenings, but a slight decline in participation rates and examination coverage. Organisational improvements were noted, yet gaps in training and equity-targeted actions remained. Conclusion: The PRECEDE–PROCEED model audits led to notable improvements in the quality and equity of cancer screening programs in Lombardy. Sustained focus on digital integration, continuous re-training, and targeted equity interventions is essential for further progress.

Published
2024
Full Text
View/download PDF

10. View-to-Label: Multi-View Consistency for Self-Supervised 3D Object Detection

Author

Mouawad, Issa, Brasch, Nikolas, Manhardt, Fabian, Tombari, Federico, and Odone, Francesca

Subjects
Computer Science - Computer Vision and Pattern Recognition
Abstract

For autonomous vehicles, driving safely is highly dependent on the capability to correctly perceive the environment in 3D space, hence the task of 3D object detection represents a fundamental aspect of perception. While 3D sensors deliver accurate metric perception, monocular approaches enjoy cost and availability advantages that are valuable in a wide range of applications. Unfortunately, training monocular methods requires a vast amount of annotated data. Interestingly, self-supervised approaches have recently been successfully applied to ease the training process and unlock access to widely available unlabelled data. While related research leverages different priors including LIDAR scans and stereo images, such priors again limit usability. Therefore, in this work, we propose a novel approach to self-supervise 3D object detection purely from RGB sequences alone, leveraging multi-view constraints and weak labels. Our experiments on KITTI 3D dataset demonstrate performance on par with state-of-the-art self-supervised methods using LIDAR scans or stereo images.

Published
2023

12. First Transcriptome Analysis of Hepatoblastoma in Brazil: Unraveling the Pivotal Role of Noncoding RNAs and Metabolic Pathways

Author

Aguiar, Talita Ferreira Marques, Rivas, Maria Prates, de Andrade Silva, Edson Mario, Pires, Sara Ferreira, Dangoni, Gustavo Dib, Macedo, Taiany Curdulino, Defelicibus, Alexandre, Barros, Bruna Durães de Figueiredo, Novak, Estela, Cristofani, Lilian Maria, Odone, Vicente, Cypriano, Monica, de Toledo, Silvia Regina Caminada, da Cunha, Isabela Werneck, da Costa, Cecilia Maria Lima, Carraro, Dirce Maria, Tojal, Israel, de Oliveira Mendes, Tiago Antonio, and Krepischi, Ana Cristina Victorino

Published
2024
Full Text
View/download PDF

14. Analysis of the direct economic impact of smoking-related hospitalizations in Italy

Author

Irene Possenti, Marco Scala, Magda Rognoni, Alessandra Lugo, Maria S. Cattaruzza, Sabrina Molinaro, Anna Odone, Luc J. M. Smits, Vincenzo Zagà, Silvano Gallus, and Luca Cavalieri d'Oro

Subjects
italy, tobacco, smoking, burden of disease, cost of illness, Diseases of the respiratory system, RC705-779, Neoplasms. Tumors. Oncology. Including cancer and carcinogens, RC254-282
Abstract

Introduction Tobacco-related diseases have a substantial economic impact in terms of medical expenses, loss of productivity, and premature death. Tobacco use is estimated to be responsible for more than 90000 deaths each year in Italy. We aimed to evaluate the annual direct economic impact on the National Health System of hospitalizations attributable to tobacco smoking in Italy. Methods We analyzed data from all the hospitalizations of patients aged ≥30 years that occurred in Italy for 12 selected tobacco-related diseases, during 2018. These diseases included oropharyngeal cancer, esophageal cancer, gastric cancer, lung cancer, pancreatic cancer, bladder cancer, laryngeal cancer, ischemic heart disease, stroke, diseases of arteries, arterioles, and capillaries, pneumonia and influenza, and chronic obstructive pulmonary disease. We obtained information on 984322 hospital discharge records, including each hospitalization's direct costs. Using relative risk estimates from the scientific literature, we computed the population attributable fraction for various tobacco-related diseases to estimate the economic impact attributable to tobacco smoking. Results One-third of all hospitalizations occurred in 2018 in Italy among people aged ≥30 years for 12 tobacco-related diseases were found to be attributable to smoking, accounting for a total cost of €1.64 billion. Among the diseases considered, those with the highest expenditures attributable to tobacco smoking were ischemic heart disease, cerebrovascular disease, and lung cancer, accounting for €556 million, €290 million, and €229 million, respectively. Conclusions Tobacco has a substantial economic impact in Italy, accounting for around 6% of the total cost of hospitalizations in 2018. This figure is expected to be largely underestimated due to several conservative assumptions adopted in the statistical analyses. It is imperative to prioritize comprehensive tobacco control measures to counteract the huge healthcare costs due to tobacco smoking.

Published
2024
Full Text
View/download PDF

15. MONOMORPHIC POST-TRANSPLATATION LYMPHOPROLIFERATIVE DISORDERS AFTER SOLID ORGAN TRANSPLANTATION – SINGLE INSTITUTE EXPERIENCE

Author

LFC Oliveira, MM Komatsu, ML Campoy, TG Cavalcante, AP Dutra, MA Monteiro, MTA Almeida, GZ Netto, LM Cristofani, and V Odone-Filho

Subjects
Diseases of the blood and blood-forming organs, RC633-647.5
Abstract

Introduction: Solid organ transplantation (SOT) is an effective treatment modality for children and young adults with end-stage organ failure with 5-year overall survival ranging from 25 to 95% depending on the transplanted organ. Patients subsequently require lifelong immunossupression to maintain graft health and limit the risk of graft rejection. Commonly used immunosuppressants to inhibit organ recipient T-lymphocyte function include calcineurin inhibitors (CNI), mTOR inhibitors, thiopurines, and antimetabolites. Most cases of monomorphic Post-Transplantation Lymphoproliferative Disorders (m-PTLD) are related to the Epstein Barr virus (EBV) and occur in patients who were EBV seronegative before the transplant. Objective: Our aim is to assess the overall survival in patients diagnosed with monomorphic Post-Transplantation Lymphoproliferative Disorders after SOT. Methods: We present retrospective data on 46 pediatric patients diagnosed with monomorphic Post-Transplantation Lymphoproliferative Disorders with Non-Hodgkin Lymphoma (NHL) after SOT in our Institution between 2001 and 2024. Results: Twenty four of the 46 were male, and 22 were female, and mean age at the time of diagnosis of PTLD was 8 years (age range, 11 months to 17 years). PTLD was diagnosed at a median time of 36 months after the organ transplant, with a significantly shorter interval in liver (20/46), 31 months, versus heart (15/46) or kidney (11/46) graft recipients (48- and 42-months years, respectively p < 0,05). Most common histopathology subtypes were Burkitt in 20/46 (44%), Diffuse Large B-cell lymphoma (DLBCL) in 19/46 (41%) and other NHL in 7/46 (15%). Most common sites were abdominal (17/46), followed by cervical and axillary lymph nodes (12/46). Murphy staging system revealed: Stage I (5/46 - 11%), Stage II (5/46 - 11%) and Stage III (33/46 -72%). Stage IV diseases with bone marrow (1/46 – 2%) and/or Central nervous system Involvement (CNS) (2/46 - 4%). The treatment carried out was reduction of immunosuppression and rituximab in 18 patients, with chemotherapy, rituximab and reduction of immunosuppression in 28 patients. In one relapse there was a need for bone marrow transplantation. No differences in outcome were observed between Early (22/46) and late-onset (24/46) PTLD and EBV-positive (14/46) or EBV-negative (25/46) PTLD, respectively. The 5-year overall and event-free survival were 91% and 83%, respectively. Conclusion: The outcome of m-PTLD has clearly improved as a result of the introduction of more uniform treatment protocols and improved supportive care, instaging and response monitoring. The majority of had a good response with chemotherapy and/or rituximab.

Published
2024
Full Text
View/download PDF

17. On the Assessment of Gray Code Kernels for Motion Characterization in People with Multiple Sclerosis: A Preliminary Study

Author

Moro, Matteo, Cellerino, Maria, Inglese, Matilde, Casadio, Maura, Odone, Francesca, Noceti, Nicoletta, Angrisani, Leopoldo, Series Editor, Arteaga, Marco, Series Editor, Chakraborty, Samarjit, Series Editor, Chen, Jiming, Series Editor, Chen, Shanben, Series Editor, Chen, Tan Kay, Series Editor, Dillmann, Rüdiger, Series Editor, Duan, Haibin, Series Editor, Ferrari, Gianluigi, Series Editor, Ferre, Manuel, Series Editor, Jabbari, Faryar, Series Editor, Jia, Limin, Series Editor, Kacprzyk, Janusz, Series Editor, Khamis, Alaa, Series Editor, Kroeger, Torsten, Series Editor, Li, Yong, Series Editor, Liang, Qilian, Series Editor, Martín, Ferran, Series Editor, Ming, Tan Cher, Series Editor, Minker, Wolfgang, Series Editor, Misra, Pradeep, Series Editor, Mukhopadhyay, Subhas, Series Editor, Ning, Cun-Zheng, Series Editor, Nishida, Toyoaki, Series Editor, Oneto, Luca, Series Editor, Panigrahi, Bijaya Ketan, Series Editor, Pascucci, Federica, Series Editor, Qin, Yong, Series Editor, Seng, Gan Woon, Series Editor, Speidel, Joachim, Series Editor, Veiga, Germano, Series Editor, Wu, Haitao, Series Editor, Zamboni, Walter, Series Editor, Zhang, Junjie James, Series Editor, Tan, Kay Chen, Series Editor, Bellotti, Francesco, editor, Grammatikakis, Miltos D., editor, Mansour, Ali, editor, Ruo Roch, Massimo, editor, Seepold, Ralf, editor, Solanas, Agusti, editor, and Berta, Riccardo, editor

Published
2024
Full Text
View/download PDF

18. Top-Tuning: a study on transfer learning for an efficient alternative to fine tuning for image classification with fast kernel methods

Author

Alfano, Paolo Didier, Pastore, Vito Paolo, Rosasco, Lorenzo, and Odone, Francesca

Subjects
Computer Science - Machine Learning
Abstract

The impressive performance of deep learning architectures is associated with a massive increase in model complexity. Millions of parameters need to be tuned, with training and inference time scaling accordingly, together with energy consumption. But is massive fine-tuning always necessary? In this paper, focusing on image classification, we consider a simple transfer learning approach exploiting pre-trained convolutional features as input for a fast-to-train kernel method. We refer to this approach as \textit{top-tuning} since only the kernel classifier is trained on the target dataset. In our study, we perform more than 3000 training processes focusing on 32 small to medium-sized target datasets, a typical situation where transfer learning is necessary. We show that the top-tuning approach provides comparable accuracy with respect to fine-tuning, with a training time between one and two orders of magnitude smaller. These results suggest that top-tuning is an effective alternative to fine-tuning in small/medium datasets, being especially useful when training time efficiency and computational resources saving are crucial.

Published
2022

19. Efficient Unsupervised Learning for Plankton Images

Author

Alfano, Paolo Didier, Rando, Marco, Letizia, Marco, Odone, Francesca, Rosasco, Lorenzo, and Pastore, Vito Paolo

Subjects
Computer Science - Computer Vision and Pattern Recognition, Computer Science - Artificial Intelligence, Computer Science - Machine Learning
Abstract

Monitoring plankton populations in situ is fundamental to preserve the aquatic ecosystem. Plankton microorganisms are in fact susceptible of minor environmental perturbations, that can reflect into consequent morphological and dynamical modifications. Nowadays, the availability of advanced automatic or semi-automatic acquisition systems has been allowing the production of an increasingly large amount of plankton image data. The adoption of machine learning algorithms to classify such data may be affected by the significant cost of manual annotation, due to both the huge quantity of acquired data and the numerosity of plankton species. To address these challenges, we propose an efficient unsupervised learning pipeline to provide accurate classification of plankton microorganisms. We build a set of image descriptors exploiting a two-step procedure. First, a Variational Autoencoder (VAE) is trained on features extracted by a pre-trained neural network. We then use the learnt latent space as image descriptor for clustering. We compare our method with state-of-the-art unsupervised approaches, where a set of pre-defined hand-crafted features is used for clustering of plankton images. The proposed pipeline outperforms the benchmark algorithms for all the plankton datasets included in our analysis, providing better image embedding properties., Comment: 13 pages. Accepted at the 26TH International Conference on Pattern Recognition (ICPR 2022)

Published
2022

24. Herpes zoster (HZ) vaccine coverage and confidence in Italy: a Nationwide cross-sectional study, the OBVIOUS project

Author

Aurelia Salussolia, Angelo Capodici, Francesca Scognamiglio, Giusy La Fauci, Giorgia Soldà, Marco Montalti, Zeno Di Valerio, Maria Pia Fantini, Anna Odone, Claudio Costantino, Heidi J. Larson, Julie Leask, Jacopo Lenzi, Davide Gori, and on behalf of the OBVIOUS board

Subjects
Herpes zoster, Vaccine, Uptake, Hesitancy, Confidence, Safety perception, Infectious and parasitic diseases, RC109-216
Abstract

Abstract Background Herpes Zoster is an age dependent disease and as such it represents a problem in the Italian social context, where the demographic curve is characterized by an overrepresentation of the elderly population. Vaccines against Herpes Zoster are available, safe and effective, however coverage remains sub-optimal. This study was therefore conducted to examine the variations in Herpes Zoster vaccine uptake and confidence across different regions in Italy. Methods This study utilized a cross-sectional computer-assisted web interview (CAWI) methodology. The survey was conducted by Dynata, an online panel provider, and involved 10,000 respondents recruited in Italy between April 11 and May 29, 2022. The sample was stratified based on geographic region, gender, and age group. Data management adhered to European Union data protection regulations, and the survey covered demographics, living conditions, and vaccination against herpes zoster (HZ), following the BeSD framework. Results The findings indicate regional disparities in herpes zoster vaccine uptake across Italy. Notably, the Islands region exhibits a particularly low vaccination rate (2.9%), highlighting the need for targeted interventions. The multivariate regression analysis showed that sociodemographic factors, limited access to healthcare services, and inadequate awareness of vaccine eligibility contribute to the lower uptake observed in this region. Conclusion In conclusion, this research emphasizes regional disparities in herpes zoster (HZ) vaccination uptake in Italy. Demographic, socioeconomic, and geographic factors impact individuals’ willingness to receive the vaccine. The study highlights the importance of awareness of vaccine eligibility and accessible vaccination facilities in increasing uptake rates.

Published
2024
Full Text
View/download PDF

25. Pneumococcal vaccine uptake among high-risk adults and children in Italy: results from the OBVIOUS project survey

Author

Zeno Di Valerio, Giusy La Fauci, Francesca Scognamiglio, Aurelia Salussolia, Marco Montalti, Angelo Capodici, Maria Pia Fantini, Anna Odone, Claudio Costantino, Giorgia Soldà, Heidi J. Larson, Julie Leask, Jacopo Lenzi, Davide Gori, and the OBVIOUS board

Subjects
Pneumococcal vaccine, Vaccine uptake, Immunization, Public aspects of medicine, RA1-1270
Abstract

Abstract Background Streptococcus pneumoniae infections, including Invasive Pneumococcal Diseases (IPDs), pose a substantial public health challenge, causing significant morbidity and mortality, especially among children and older adults. Vaccination campaigns have played a vital role in reducing pneumococcal-related deaths. However, obstacles related to accessibility and awareness might impede optimal vaccine adoption. This study aims to provide comprehensive data on pneumococcal vaccine coverage and attitudes within at-risk groups in Italy, with the goal of informing public health strategies and addressing vaccination barriers. Methods Between April 11 and May 29, 2022, a questionnaire investigating vaccine uptake and attitudes toward several vaccinations was administered to 10,000 Italian adults, chosen through population-based sampling. Respondents who were targets of the campaign according to the 2017–2019 National Vaccination Plan, accessed questions regarding pneumococcal vaccination. Data on uptake, awareness of having the right to free vaccination, opinion on vaccine safety, concern with pneumococcal disease, and ease of access to vaccination services were summarized and presented based on statistical regions. Multinomial logistic regression analysis was used to explore factors influencing vaccine uptake. Results Out of 2357 eligible adult respondents (42.6% women; mean age: 58.1 ± 15.7), 39.5% received pneumococcal vaccination. Uptake differed among at-risk groups: respondents aged ≥65 (33.7%), with lung disease (48.4%), cardiovascular disease (46.6%), and diabetes (53.7%). Predictors of not being vaccinated and unwilling to included female gender, residing in rural areas, lower education, low concern about pneumococcal disease, vaccine safety concerns, and associations with vaccine-opposed acquaintances. Health access issues predicted willingness to be vaccinated despite non-vaccination. Pneumopathy, heart disease, diabetes, and living in Northeastern or Central Italy were linked to higher uptake. Among the 1064 parents of eligible children, uptake was 79.1%. Parental unawareness of children’s free vaccination eligibility was a predictor of non-vaccination. Vaccine safety concerns correlated with reluctance to vaccinate children, while perceived healthcare access challenges were associated with wanting but not having received vaccination. Conclusions Pneumococcal vaccination uptake within prioritized groups and children in Italy remains inadequate. Scarce awareness of vaccine availability and obstacles in accessing vaccinations emerge as principal barriers influencing this scenario.

Published
2024
Full Text
View/download PDF

26. Video slot machine use in adolescence: the role of self-efficacy beliefs, current and expected personal fulfillment at the social and educational level

Author

Giansanto Mosconi, Joseph DelFerro, Andrea Jin, Paola Bertuccio, Anna Odone, Ilaria Albertin, Andrea Amerio, Lorella Cecconami, Marcello Esposito, Simone Feder, Silvano Gallus, Sabrina Molinaro, Anna Polgatti, Sara Russo, Franco Taverna, Diego Turcinovich, and Tomaso Vecchi

Subjects
Gambling, Slot machine, Adolescents, Cross-sectional study, Psychology, Associated factors, BF1-990, Social pathology. Social and public welfare. Criminology, HV1-9960
Abstract

Introduction: Video slot machines (VSMs), which are available on electronic gaming machines and online platforms, are widely recognized for their high potential to trigger harmful gambling behaviors. Yet, the psychological factors associated with VSM use among adolescents remain poorly investigated. Methods: We estimated adjusted odds ratios (OR) and corresponding 95% confidence intervals (CI) of current regular (at least monthly) VSM use according to self-efficacy beliefs regarding everyday problem-solving ability, along with current and expected personal fulfillment at the social, and educational level among 7,959 Italian high school students aged 15–17. Results: Current regular VSM use was reported by 108 students (1.4%). Notably, this behavior was more frequently reported by those with either particularly high or low self-ratings in problem-solving ability, current and expected social fulfillment, as well as by those with lower self-ratings in both current and expected educational/work fulfillment. Conclusions: These findings underscore that during adolescence, gambling behaviors are associated with diverse psychological profiles, each entailing distinct vulnerabilities. Effectively addressing adolescent gambling necessitates a comprehensive approach, with a strong emphasis on educational support. Given the increasing challenges in curbing underage exposure and access to the most harmful gambling formats, this study offers valuable insights for the design of health promotion and prevention strategies.

Published
2024
Full Text
View/download PDF

27. Who is (not) vaccinated? A proposal for a comprehensive immunization information system

Author

Giacomo Pietro Vigezzi, Elena Maggioni, Fabrizio Bert, Corrado de Vito, Roberta Siliquini, and Anna Odone

Subjects
Immunization information system, proactive medicine, immunization programs, vaccination hesitancy, ethical issues, Immunologic diseases. Allergy, RC581-607, Therapeutics. Pharmacology, RM1-950
Abstract

The role of immunization in public health is crucial, offering widespread protection against infectious diseases and underpinning societal well-being. However, achieving optimal vaccination coverage is impeded by vaccine hesitancy, a significant challenge that necessitates comprehensive strategies to understand and mitigate its effects. We propose the integration of Population Health Management principles with Immunization Information Systems (IISs) to address vaccine hesitancy more effectively. Our approach leverages systematic health determinants analysis to identify at-risk populations and tailor interventions, thereby promoting vaccination coverage and public health responses. We call for the development of an enhanced version of the Italian National Vaccination Registry, which aims to facilitate real-time tracking of individuals’ vaccination status while improving data accuracy and interoperability among healthcare systems. This registry is designed to overcome current barriers by ensuring robust data protection, addressing cultural and organizational challenges, and integrating behavioral insights to foster informed public health campaigns. Our proposal aligns with the Italian National Vaccination Prevention Plan 2023–2025 and emphasizes proactive, evidence-based strategies to increase vaccination uptake and contrast the spread of vaccine-preventable diseases. The ultimate goal is to establish a data-driven, ethically sound framework that enhances public health outcomes and addresses the complexities of vaccine hesitancy within the Italian context and beyond.

Published
2024
Full Text
View/download PDF

29. A novel smart fertigation system for irrigation with treated wastewater: Effects on nutrient recovery, crop and soil

Author

Giordano Odone, Giulio Demetrio Perulli, Giuseppe Mancuso, Stevo Lavrnić, and Attilio Toscano

Subjects
Water scarcity, Tertiary treated wastewater, Wastewater reuse, Fertilizer saving, Plant nutritional status, Plant physiological performances, Agriculture (General), S1-972, Agricultural industries, HD9000-9495
Abstract

Both southern and northern regions of Italy are experiencing reduced precipitation and increased heat waves due to climate change, negatively affecting agricultural sector. Urban wastewater could be a solution to this problem, providing a constant source for irrigation and reducing synthetic fertilizer use. This research presents a two-year field study on using tertiary treated wastewater for processing tomato crop irrigation through an innovative smart fertigation system, designed to supply the exact doses of NPK nutrients considering those already delivered to the plants via the irrigation water. With the aim of studying the effects of irrigation with treated wastewater, three water sources were compared: fresh water added with chemical fertilizer, tertiary treated wastewater added with chemical fertilizer and tertiary treated wastewater without addition of fertilizer. The proposed system was efficient and consistent with the design, it saved considerable amounts of fertilizers, handling nutrient fluctuations in wastewater. Of the three irrigation water types that have been tested, only the one that used tertiary treated wastewater alone without additional fertilizers was not capable of meeting tomato nutritional needs, despite the fact that significant macronutrient savings were achieved. No negative effects on soil or plant physiological performances were observed. Plants irrigated with wastewater showed similar growth and productivity to those irrigated with fresh water and no significant differences in fruit quality were found, highlighting the benefits of wastewater reuse for crop irrigation.

Published
2024
Full Text
View/download PDF

30. FasterVideo: Efficient Online Joint Object Detection And Tracking

Author

Mouawad, Issa and Odone, Francesca

Subjects
Computer Science - Computer Vision and Pattern Recognition
Abstract

Object detection and tracking in videos represent essential and computationally demanding building blocks for current and future visual perception systems. In order to reduce the efficiency gap between available methods and computational requirements of real-world applications, we propose to re-think one of the most successful methods for image object detection, Faster R-CNN, and extend it to the video domain. Specifically, we extend the detection framework to learn instance-level embeddings which prove beneficial for data association and re-identification purposes. Focusing on the computational aspects of detection and tracking, our proposed method reaches a very high computational efficiency necessary for relevant applications, while still managing to compete with recent and state-of-the-art methods as shown in the experiments we conduct on standard object tracking benchmarks, Comment: Accepted at 21st International Conference on Image Analysis and Processing (ICIAP 2021)

Published
2022
Full Text
View/download PDF

31. Estimating SARS-CoV-2 transmission in educational settings: a retrospective cohort study

Author

Manica, Mattia, Poletti, Piero, Deandrea, Silvia, Mosconi, Giansanto, Ancarani, Cinzia, Lodola, Silvia, Guzzetta, Giorgio, d'Andrea, Valeria, Marziano, Valentina, Zardini, Agnese, Trentini, Filippo, Odone, Anna, Tirani, Marcello, Ajelli, Marco, and Merler, Stefano

Subjects
Physics - Physics and Society, Quantitative Biology - Populations and Evolution, Quantitative Biology - Quantitative Methods
Abstract

Background School closures and distance learning have been extensively applied to control SARS-CoV-2 transmission. Despite evidence of viral circulation in schools, the contribution of students and of in-person schooling to the transmission remains poorly quantified. Methods We analyze 976 exposure events, involving 460 positive individuals, as identified in early 2021 by routine surveillance and through an extensive screening conducted on students, school personnel, and their household members during an outbreak in a small municipality of Italy. Results From the analysis of potential transmission chains, we estimated that, on average, 55.1%, 17.3% and 27.6% infection episodes were linked to household, school, and community contacts, respectively. Clusters originated from students or school personnel showed a larger average cluster size (3.32 vs 1.15), a larger average number of generations in the transmission chain (1.56 vs 1.17) and a larger set of associated close contacts (11.3 vs 3.15, on average). We found substantial transmission heterogeneities, with 20% positive individuals seeding 75-80 of all transmission. A higher proportion of infected individuals causing onward transmission was found among students (48.8% vs 29.9%, on average), who also caused a markedly higher number of secondary cases (mean: 1.3 vs 0.5). Conclusions Uncontrolled transmission at school could disrupt the regular conduct of teaching activities, likely seeding the transmission into other settings, and increasing the burden on contact-tracing operations.

Published
2022

32. Time-to-Label: Temporal Consistency for Self-Supervised Monocular 3D Object Detection

Author

Mouawad, Issa, Brasch, Nikolas, Manhardt, Fabian, Tombari, Federico, and Odone, Francesca

Subjects
Computer Science - Computer Vision and Pattern Recognition, Computer Science - Robotics
Abstract

Monocular 3D object detection continues to attract attention due to the cost benefits and wider availability of RGB cameras. Despite the recent advances and the ability to acquire data at scale, annotation cost and complexity still limit the size of 3D object detection datasets in the supervised settings. Self-supervised methods, on the other hand, aim at training deep networks relying on pretext tasks or various consistency constraints. Moreover, other 3D perception tasks (such as depth estimation) have shown the benefits of temporal priors as a self-supervision signal. In this work, we argue that the temporal consistency on the level of object poses, provides an important supervision signal given the strong prior on physical motion. Specifically, we propose a self-supervised loss which uses this consistency, in addition to render-and-compare losses, to refine noisy pose predictions and derive high-quality pseudo labels. To assess the effectiveness of the proposed method, we finetune a synthetically trained monocular 3D object detection model using the pseudo-labels that we generated on real data. Evaluation on the standard KITTI3D benchmark demonstrates that our method reaches competitive performance compared to other monocular self-supervised and supervised methods.

Published
2022
Full Text
View/download PDF

33. Correction: First Transcriptome Analysis of Hepatoblastoma in Brazil: Unraveling the Pivotal Role of Noncoding RNAs and Metabolic Pathways

Author

Aguiar, Talita Ferreira Marques, Rivas, Maria Prates, de Andrade Silva, Edson Mario, Pires, Sara Ferreira, Dangoni, Gustavo Dib, Macedo, Taiany Curdulino, Defelicibus, Alexandre, Barros, Bruna Durães de Figueiredo, Novak, Estela, Cristofani, Lilian Maria, Odone, Vicente, Cypriano, Monica, de Toledo, Silvia Regina Caminada, da Cunha, Isabela Werneck, da Costa, Cecilia Maria Lima, Carraro, Dirce Maria, Tojal, Israel, de Oliveira Mendes, Tiago Antônio, and Krepischi, Ana Cristina Victorino

Published
2024
Full Text
View/download PDF

35. Vaccine Mandates in the COVID-19 Era: Changing Paradigm or Public Health Opportunity?; Comment on 'Convergence on Coercion: Functional and Political Pressures as Drivers of Global Childhood Vaccine Mandates'

Author

Anna Odone, Giulia Dallagiacoma, and Giacomo Pietro Vigezzi

Subjects
vaccine coverage, mandatory immunisation, covid-19 pandemic, risk perception, public health, Public aspects of medicine, RA1-1270
Abstract

The debate around vaccine mandates has flourished over the last decade, with several countries introducing or extending mandatory childhood vaccinations. In a recent study, Attwell and Hannah explore how functional and political pressures added to public health threats in selected countries, motivating governments to increase the coerciveness of their childhood vaccine regimes. In this commentary, we reflect on whether such model applies to the coronavirus disease 2019 (COVID-19) case and how the pandemic has re-shuffled the deck around vaccine mandates. We identify COVID-19 immunisation policies’ distinctive aspects as we make the case of countries implementing mass immunisation programmes while relying on digital COVID-19 certificates as an indirect form of mandate to increase vaccine uptake. We conclude by acknowledging that different forms of mandatory vaccination might serve as a shortcut to protect population health in times of emergency, underlining, however, that the ultimate public health goal is to promote voluntary, informed, and responsible adherence to preventive behaviours.

Published
2023
Full Text
View/download PDF

36. HHP-Net: A light Heteroscedastic neural network for Head Pose estimation with uncertainty

Author

Cantarini, Giorgio, Tomenotti, Federico Figari, Noceti, Nicoletta, and Odone, Francesca

Subjects
Computer Science - Computer Vision and Pattern Recognition
Abstract

In this paper we introduce a novel method to estimate the head pose of people in single images starting from a small set of head keypoints. To this purpose, we propose a regression model that exploits keypoints computed automatically by 2D pose estimation algorithms and outputs the head pose represented by yaw, pitch, and roll. Our model is simple to implement and more efficient with respect to the state of the art -- faster in inference and smaller in terms of memory occupancy -- with comparable accuracy. Our method also provides a measure of the heteroscedastic uncertainties associated with the three angles, through an appropriately designed loss function; we show there is a correlation between error and uncertainty values, thus this extra source of information may be used in subsequent computational steps. As an example application, we address social interaction analysis in images: we propose an algorithm for a quantitative estimation of the level of interaction between people, starting from their head poses and reasoning on their mutual positions. The code is available at https://github.com/cantarinigiorgio/HHP-Net., Comment: Accepted at WACV 2022

Published
2021

38. Beyond Clinical Trials: Understanding Neurotrophic Tropomyosin Receptor Kinase Inhibitor Challenges and Efficacy in Real-World Pediatric Oncology

Author

Vince, Carolina Sgarioni Camargo, Brassesco, Maria Sol, Mançano, Bruna Minniti, Gregianin, Lauro Jose, Carbone, Edna Kakitani, do Amaral e Castro, Adham, Dwan, Viviane Sayuri Yamachira, Menezes da Silva, Roberta Zeppini, Mariano, Cassia Silvestre, da Mata, Juliana França, Silva, Marcelo Oliveira, Caran, Eliana Maria Monteiro, Macedo, Carla Donato, Alves da Costa, Gildene, Esteves, Tereza Cristina, Silva, Luciana Nunes, Ferman, Sima Esther, Martins, Flavia Delgado, Cristófani, Lilian Maria, Odone-Filho, Vicente, Silva, Marcelo Milone, Reis, Rui Manuel, Pianovski, Mara Albonei Dudeque, Campregher, Paulo Vidal, Kunii, Mayara Satsuki, de Sá Rodrigues, Karla Emilia, Carvalho Filho, Neviçolino Pereira, and Valera, Elvis Terci

Published
2024
Full Text
View/download PDF

43. Global trends in youth suicide from 1990 to 2020: an analysis of data from the WHO mortality databaseResearch in context

Author

Paola Bertuccio, Andrea Amerio, Enrico Grande, Carlo La Vecchia, Alessandra Costanza, Andrea Aguglia, Isabella Berardelli, Gianluca Serafini, Mario Amore, Maurizio Pompili, and Anna Odone

Subjects
Suicide, Mortality, Official statistics, Global, Trends, Youth, Medicine (General), R5-920
Abstract

Summary: Background: Suicide is a serious but preventable public health concern at the global level, showing relevant geographical differences. This study aims to monitor global temporal and geographical patterns in suicide mortality in pre-adolescents, adolescents, and young adults (i.e., aged 10–24 years), from 1990 to 2020 or the most recent available year. Methods: Using the World Health Organisation mortality database, we conducted an analysis on a subset of 52 countries with valid and high-quality data. We computed age-standardised suicide rates (ASR) by sex, country, and calendar year, and performed a joinpoint regression analysis to identify significant changes in the temporal suicide trends over the studied period. Findings: High variability in suicide rates and trends was observed, with a male-to-female ratio of two to five. Between 1990 and 2020, most European countries reported declining suicide trends, with some exceptions. In particular, alarming trends emerged in the United Kingdom, with annual rises of 2.5% (95% CI: 1.6–3.5) since 2005 among males and 8.5% (95% CI: 4.7–12.6) since 2012 among females. The most favorable trends and lowest suicide rates were in Southern Europe, with 3.1/100,000 persons in Italy (2020) and 3.5/100,000 persons in Spain (2021) among males, and 0.9/100,000 persons in Italy (2020) and 1.1/100,000 persons in Romania (2019) among females. Conversely, the highest rates were in Central-Eastern Europe, with 10.2/100,000 males in the Russian Federation (2019) and 10.0/100,000 males in Poland (2002). Higher suicide rates and significant increases were reported in not European areas. The highest ASR was 15.5/100,000 males in the United States of America, with an annual increase of 3.8% (95% CI: 3.1–4.5) among males in 2009–2020 and 6.7% (95% CI: 5.6–7.8) among females in 2007–2017, followed by a levelling off. Interpretation: Temporal and geographical comparisons of suicide mortality should be interpreted with caution due to potential misclassification or under-reporting of suicide deaths in some countries. Funding: None.

Published
2024
Full Text
View/download PDF

Catalog

Books, media, physical & digital resources
See catalog results