Navegando por Data de Publicação, começando com "2025-10-29"

Filtrar resultados por ano ou mês

Agora exibindo 1 - 3 de 3

Multi-pathology segmentation in lumbar spine MRI: a comparative deep learning approach
(Universidade Federal de São Carlos, 2025-10-29) Leite Junior, Claudio Luiz; Almeida Junior, Jurandy Gomes de; https://lattes.cnpq.br/4495269939725770; https://orcid.org/0000-0002-4998-6996; https://lattes.cnpq.br/8623721713277640; https://orcid.org/0009-0001-2405-0075; Comin, Cesar Henrique; Cappabianco, Fabio Augusto Menocci; https://lattes.cnpq.br/9563440403120931; https://lattes.cnpq.br/7438076121387151; https://orcid.org/0000-0003-1207-4982; https://orcid.org/0000-0002-2139-7938
Low back pain is a leading cause of disability worldwide, and Magnetic Resonance Imaging (MRI) of the lumbar spine is fundamental for its diagnosis. However, the manual analysis of these images is a time-consuming and subjective process, and existing computational methods are often limited to a single pathology or rely on complex, multi-stage pipelines. A central challenge, not yet fully addressed, is the simultaneous occurrence of multiple pathologies in the same anatomical structure—a common clinical scenario that current models fail to model effectively. This dissertation addresses this gap by proposing and validating a robust methodology for the automated segmentation of multiple co-existing pathologies in lumbar intervertebral discs. The work was structured in two complementary phases. The first established the methodological underpinnings of this research. It consisted of a rigorous empirical study evaluating five deep learning architectures and four loss functions to determine the most effective approaches for the fundamental task of segmenting the vertebrae and intervertebral discs. This step ensured that the subsequent investigation into pathologies was built upon a robust and validated base. The second phase, the main contribution of this work, systematically investigated three distinct strategies for multi-pathology segmentation: (i) binary class segmentation, a baseline that treats each pathology independently; (ii) multi-class segmentation, mapping 70 disease combinations to unique classes (non-overlapping masks); and (iii) multi-label segmentation, which uses binary channels to explicitly model the coexistence of multiple diagnoses (overlapping masks). Our results, derived from over 200 training pipelines, demonstrate that the multi-label approach, especially when implemented with the V-Net and Swin UNETR architectures, achieves diagnostic accuracy comparable to the baseline while offering significantly superior computational efficiency. By developing a unified framework that integrates the precise spatial localization of symptomatic areas with the classification of multiple concurrent diseases, this work establishes a practical and efficient guideline for future research and clinical applications in the automated diagnosis of spinal pathologies.
Projeto de rede de transporte público urbano: uma abordagem baseada em otimização multiobjetivo da infraestrutura prioritária e eletrificação de frota de ônibus
(Universidade Federal de São Carlos, 2025-10-29) Borchers, Tatiane; Fernandes, Ricardo Augusto Souza; https://lattes.cnpq.br/0880243208789454; https://lattes.cnpq.br/5093971220777490; https://orcid.org/0000-0002-0186-6661; Almeida, Madson Cortes de; Rabelo, Ricardo de Andrade Lira; Sanches, Danilo Sipoli; Borges, Fábbio Anderson Silva; https://lattes.cnpq.br/7592852392858067; https://lattes.cnpq.br/2263553634422386; https://lattes.cnpq.br/6377657274398145; https://lattes.cnpq.br/5378006986523567
In the context of the climate emergency, where cities have a significant impact on greenhouse gas emissions, the transport sector is one of the main contributors to both emissions and the depletion of non-renewable resources. This thesis presents an analysis of the impact of integrating sustainability strategies into bus public transport systems to reduce carbon dioxide (CO2), particulate matter (PM), and nitrogen oxide (NOx) emissions. The modelling approach considered the perspectives of passengers, operators, and the community within the transport planning process, while simultaneously addressing environmental, economic, and social sustainability aspects. A multi-objective optimisation model was developed for the Urban Transport Network Design Problem, whose decision variables were the implementation of exclusive bus lanes and the electrification of bus fleets. The objective functions aimed to minimise the construction, acquisition, maintenance, and operation costs, as well as the construction and operational emissions of a bus public transport system. The European Environment Agency's model was used for the emission modelling formulation; it takes into account the technology and category of the vehicles and their operating speed, the latter being a linking parameter between the two decision variables. To solve the mathematical model, the metaheuristic NSGA-II (Non-dominated Sorting Genetic Algorithm II) was employed, which was applied to the city of Belo Horizonte, Minas Gerais, Brazil. The case study was defined based on a qualitative analysis of principles and an investigation of the availability of open data in Brazilian state capitals. The results indicated that the strategies proposed are capable of improving the quality of public transport systems by reducing CO2, PM and NOx emissions. The scenario analysis demonstrated that when the measures were applied jointly, using overnight and opportunity charging during the day, the costs, although high, yielded greater environmental and social benefits for passengers and the community, with reductions in operational emissions compared to the baseline scenario of approximately 39.1 to 57.8% for CO₂, 92.9 to 94.9% for PM, and 86.3 to 90.5% for NOx. As contributions, this thesis sought to overcome the gap between mathematical complexity and application complexity by using an emission model dependent on bus technology, category, and operating speed, and by applying it to a large-scale network.
Satisfação e padrão de sangramento de mulheres férteis em situação de vulnerabilidade usuárias do implante de etonogestrel e uma reflexão sobre a importância da iniciação científica na formação médica de uma estudante de medicina
(Universidade Federal de São Carlos, 2025-10-29) Petrocheli, Bruna Beatriz; Nasser, Michel; https://lattes.cnpq.br/1629054550833348; https://orcid.org/0000-0003-3087-9784; https://lattes.cnpq.br/3083360352564648; https://orcid.org/0000-0001-9853-5131
In a context where more than half of pregnancies in Brazil are unplanned—a recognized public health issue—and considering that women in situations of vulnerability are more susceptible to the negative impacts of such events, the adoption of contraceptive methods that align with the realities of this group is an important strategy to prevent further increases in these numbers and to mitigate the complications of a potential high-risk pregnancy. This project aims to evaluate the satisfaction level and bleeding patterns among users of the etonogestrel implant in a municipality in the interior of São Paulo state. The study has an observational, cross sectional, descriptive/quantitative design and targets women of reproductive age in vulnerable situations who have been using the method for at least three months. The most prevalent age group was 26–35 years (39.6%), and the majority of participants self-identified as mixed race or white (93%). Most women had previously used some form of contraception (79%), with combined oral contraceptives and the quarterly injectable being the most common. Satisfaction with the method was considered high (64.9% reported being very satisfied), with the main positive aspects being its efficacy (71.9%) and convenience (64.9%). Bleeding patterns between 3 and 6 months of use were predominantly unfavorable (57.2%), but tended to improve to a favorable pattern (77.8%) after 12 months. Thus, the etonogestrel implant proved to be a well-accepted contraceptive method with notable advantages among women in vulnerable situations, reinforcing the importance of its availability for this population and the need to expand access to it as a contraceptive option within the public healthcare system.