Navegando por Palavras-chave "Healthcare"
Agora exibindo 1 - 4 de 4
Resultados por página
Opções de Ordenação
- ItemAcesso aberto (Open Access)Construção de interface gráfica para consolidação e análise de dados resultante de exames médicos(Universidade Federal de São Paulo, 2023-07-01) Mendonça, Grazieli Leite; Polli, Roberson Saraiva [UNIFESP]; http://lattes.cnpq.br/6933632737044678A análise de dados tem se tornado cada vez mais importante na área da saúde, desempenhando um papel crucial na identificação de padrões, detecção de anomalias e tomada de decisões clínicas. Para realizar essas análises, o principal desafio tem se tornado lidar com arquivos volumosos que contêm muitas vezes resultados de milhões de exames por mês, e esse é o desafio enfrentado pela equipe de assessores da empresa Siemens Healthineers. Com o objetivo de solucionar este problema, foi desenvolvida uma ferramenta que permite unir diversos arquivos no formato .csv em um único arquivo, facilitando a análise dos dados por meio de filtragem e cálculos estatísticos. Além disso, com o objetivo de trazer mais facilidade para a equipe da Siemens Healthineers, a ferramenta também realiza cálculos estatístico e retorna tabelas e gráficos para analisar os resultados de uma forma mais rápida, tudo isso com uma interface gráfica que foi pensada na melhor experiência do usuário. Os resultados mostraram que a ferramenta atendeu às necessidades dos assessores, proporcionando uma análise mais eficiente e fornecendo insights valiosos para melhorar a forma como é abordado o desempenho dos equipamentos, de forma a garantir que o cliente receba o melhor desempenho do equipamento.
- ItemSomente MetadadadosEACTS in the future: second strategic conference. the view from the BRICS countries(Oxford Univ Press Inc, 2013-01-01) Gomes, Walter J. [UNIFESP]; Universidade Federal de São Paulo (UNIFESP)BRICS is an acronym for Brazil, Russia, India, China and South Africa and has emerged as the symbol of the shift in global economic power, developing at a faster pace than industrialized countries. BRICS accounted for 53% of the entire global GDP growth during the period 2007-2010 and, in the next 40 years, as much as 80% of the world's economic growth will come from emerging market countries. Despite the fact that infrastructure in BRICS has improved markedly in recent years, these countries have not created a modern, broad healthcare system as encountered in the G7 industrialized countries and extensive regional differences in health expenditure exist between them. Nevertheless, the BRICS countries are quickly taking the lead in encouraging innovation, simplifying devices and processes and applying newer technologies that are more adapted to consumers' needs and less costly. Cardiovascular surgery in the BRICS countries remains far lower when compared with the G7 countries and the cardiovascular surgical training also varies widely. However, this huge shift in the global economy and the regional discrepancies might represent a unique opportunity for co-operation, interaction and partnership to integrate cardiovascular societies and surgeons all over the globe for the best care of our patients: surely it will contribute to making our world more egalitarian, fairer and better.
- ItemAcesso aberto (Open Access)Machine learning for healthcare: a data-centric approach(Universidade Federal de São Paulo, 2024-06-25) Valeriano, Maria Gabriela [UNIFESP]; Lorena, Ana Carolina; Kiffer, Carlos Roberto Veiga [UNIFESP]; http://lattes.cnpq.br/7021893874375037; http://lattes.cnpq.br/3451628262694747; http://lattes.cnpq.br/7462488231975857Machine learning models have the potential to revolutionize the healthcare sector by leveraging continuously collected data in health systems. Traditionally, these models are trained on large datasets, with performance improvements achieved through robust models and hyperparameter tuning. In this work, we propose a data-centric approach focusing on improving the data itself. Throughout this research, a set of health-related databases was created. These databases originate from four distinct sources, encompassing the prediction of severe cases of COVID-19 and dengue, as well as the authorization of specialized care in the public health system in Brazil. The datasets created cover seven predictive tasks, each with separate training and testing data. All problems were designed as binary classification tasks and adopted tabular data. The datasets were initially characterized in relation to their hardness profiles, using a specific hardness measure proposed in previous works. This measure considers the probability of an instance being misclassified by different machine learning algorithms. Our analysis considered seven classifiers with distinct biases: Gradient Boosting, Random Forest, Logistic Regression, Multilayer Perceptron, Support Vector Classifier (with linear and RBF kernels), and Bagging. The models were evaluated using a set of metrics, area under the ROC curve and per-class recall and precision, to provide a holistic consideration of model performance. We proposed a new approach to generate post-hoc explanations for machine learning models. In this approach, we identified instances where the models are most likely to fail, offering data-centric explanations for such failures. The patterns found explain the model errors, resulting in greater confidence in the predictions made. Additionally, we present a case study where instance hardness analysis was adopted to improve the design of a prediction problem in collaboration with the data specialist. Our work demonstrated that through this approach, it was possible to improve data quality and, ultimately, model performance. Finally, we propose a generalized approach to enhance model performance when access to data experts is not possible. A two-step strategy was adopted: first, cleaning the training data based on instance difficulty values, and then introducing a reject option when the models did not offer high-confidence predictions for test instances. The results show that it is possible to improve model performance at the cost of rejecting instances from the test set.
- ItemAcesso aberto (Open Access)Profissionais de saúde e violência intrafamiliar contra a criança e adolescente(Escola Paulista de Enfermagem, Universidade Federal de São Paulo (UNIFESP), 2009-01-01) Nunes, Cristina Brandt; Sarti, Cynthia Andersen [UNIFESP]; Ohara, Conceicao Vieira da Silva [UNIFESP]; Universidade Federal de Mato Grosso do Sul; Universidade Federal de São Paulo (UNIFESP)OBJECTIVES: To understand how heath care professionals approach family violence against children and teenagers. METHODS: This was a qualitative case study with 30 health care professionals. RESULTS: Health care professionals were concerned with the lack of successful family problems resolution. Measures used by health care professionals emphasized punitive actions instead of caring behaviors. The characteristics of the job did not allow of the health care professionals to express their feelings and reactions and to know how to successfully address family violence. Health care professionals' approaches to address violence with families who already experienced violence may also become violent acts against those families. CONCLUSION: Approaches used to address family violence against children and teenagers reflect a lack of integration among the several categories of health care professionals and health care services.