Machine Learning na biomedicina: breve revisão bibliográfica
Nenhuma Miniatura disponível
Data
2023-09-11
Tipo de documento
Artigo Científico
Título da Revista
ISSN da Revista
Título de Volume
Área do conhecimento
Ciências da Saúde
Modalidade de acesso
Acesso aberto
Editora
Autores
Marcondes, Michaela Shiotani
Lima, Luan Gonçalves de
Gonzaga, Mirielle de Cássia
Oliveira, Lívia Sanches de
Orientador
Santos, Joyce Nascimento
Coorientador
Resumo
As ciências biológicas e da saúde são áreas em constante evolução que ocupam lugar de vanguarda na pesquisa e no avanço tecnológico e que se utilizam de recursos que demandam alto poder numérico para melhoria da qualidade e velocidade de processamento de dados em diversos setores. Dentre a subárea da Inteligência Artificial, destaca-se o Machine Learning como ferramenta computacional essencial em estudos clínicos na área biomédica, atribuindo
vantagens em análises preditivas e prognósticas. Contudo, sua finalidade não é apenas representar e predizer dados, mas gerar protótipos que possam ser interpretados e ensinados com base em algoritmos consolidados para aporte na prática clínica. Diante, da análise de dados complexos, relatórios e imagens que demandam tempo e empenho dos profissionais da saúde, o emprego do ML agrega eficiência ao processo e permite que informações sejam analisadas precisamente em um menor intervalo de tempo, além de identificar padrões complexos em largos conjuntos de dados médicos, facultando a estes profissionais uma tomada de decisão mais informada e precisa. Logo, áreas como diagnóstico por imagem, oncologia, cardiologia e citopatologia, estão sendo beneficiadas por técnicas aplicadas neste campo de estudo. Ademais, utilizadas na vigilância de patologias, através de sensores vestíveis e na modelagem preditiva na saúde, possibilitando a utilização de dados e algoritmos estatísticos para estimar resultados futuros, de acordo com os dados históricos de cada indivíduo. Portanto, a aplicação de ML na área da saúde tem um enorme potencial para transformar a forma como os especialistas lidam com diagnósticos e tomada de decisões, visto que, a capacidade de processar grandes quantidades de dados biológicos permite que a inteligência artificial forneça insights precisos para ajudar a identificar e tratar doenças de forma mais eficaz e evitar erros comuns na área da saúde. Desta forma, o objetivo do presente estudo é analisar de que forma a ML trará benefícios práticos para área biomédica a partir da assertividade dos estudos clínicos, da precisão do diagnóstico e da eficácia no tratamento personalizado, visto que, a subjetividade e o retorno tardio do diagnóstico requerem novas metodologias mais consistentes
Biological and health sciences are areas in constant evolution that occupy a vanguard position in research and technological advancement and that use resources that demand high numerical power toimprove the quality and speed of data processing in several sectors. Within the subarea of Artificial Intelligence, Machine Learning stands out as an essential computational tool in clinical studies in the biomedical area, attributing advantages in predictive and prognostic analysis. However, its purpose is not only to represent and predict data, but to generate prototypes that can be interpreted and taught based on consolidated algorithms to support clinical practice. Faced with the analysis of complex data, reports and images that demand time and commitment from health professionals, the use of ML adds efficiency to the process and allows information to be analyzed precisely in a shorter period of time, in addition to identifying complex patterns in largemedical datasets, providing these professionals with more informed and accurate decision-making. Therefore, areas such as diagnostic imaging, oncology, cardiology and cytopathology are benefiting from techniques applied in this field of study. Furthermore, they are used in disease surveillance, through wearable sensors and in predictive modeling in health, enabling the use of data and statistical algorithms to estimate future results, according to the historical data of each individual. Therefore, the application of ML in healthcare has enormous potential to transform the way specialists deal with diagnoses and decision-making, since the ability to process large amounts of biological data allows artificial intelligence to provide accurate insights for help identify and treat illnesses more effectively and avoid common mistakes in healthcare. In this way, the objective of the present study is to analyze how LM will bring practical benefits to the biomedical area based on the assertiveness of clinical studies,the accuracy of the diagnosis and the effectiveness in personalized treatment, since the subjectivity and the late return of the diagnosis require new, more consistent methodologies.
Biological and health sciences are areas in constant evolution that occupy a vanguard position in research and technological advancement and that use resources that demand high numerical power toimprove the quality and speed of data processing in several sectors. Within the subarea of Artificial Intelligence, Machine Learning stands out as an essential computational tool in clinical studies in the biomedical area, attributing advantages in predictive and prognostic analysis. However, its purpose is not only to represent and predict data, but to generate prototypes that can be interpreted and taught based on consolidated algorithms to support clinical practice. Faced with the analysis of complex data, reports and images that demand time and commitment from health professionals, the use of ML adds efficiency to the process and allows information to be analyzed precisely in a shorter period of time, in addition to identifying complex patterns in largemedical datasets, providing these professionals with more informed and accurate decision-making. Therefore, areas such as diagnostic imaging, oncology, cardiology and cytopathology are benefiting from techniques applied in this field of study. Furthermore, they are used in disease surveillance, through wearable sensors and in predictive modeling in health, enabling the use of data and statistical algorithms to estimate future results, according to the historical data of each individual. Therefore, the application of ML in healthcare has enormous potential to transform the way specialists deal with diagnoses and decision-making, since the ability to process large amounts of biological data allows artificial intelligence to provide accurate insights for help identify and treat illnesses more effectively and avoid common mistakes in healthcare. In this way, the objective of the present study is to analyze how LM will bring practical benefits to the biomedical area based on the assertiveness of clinical studies,the accuracy of the diagnosis and the effectiveness in personalized treatment, since the subjectivity and the late return of the diagnosis require new, more consistent methodologies.
Palavras-chave
Inteligência artificial, Machine learning, Diagnóstico por imagem