Publicação
Data mining predictive models for pervasive intelligent decision support in intensive care medicine
| Resumo: | The introduction of an Intelligent Decision Support System (IDSS) in a critical area like the Intensive Medicine is a complex and difficult process. In this area, their professionals don’t have much time to document the cases, because the patient direct care is always first. With the objective to reduce significantly the manual records and, enabling, at the same time, the possibility of developing an IDSS which can help in the decision making process, all data acquisition process and knowledge discovery in database phases were automated. From the data acquisition to the knowledge discovering, the entire process is autonomous and executed in real-time. On-line induced data mining models were used to predict organ failure and outcome. Preliminary results obtained with a limited population of patients showed that this approach can be applied successfully. |
|---|---|
| Autores principais: | Portela, Filipe |
| Outros Autores: | Pinto, Filipe; Santos, Manuel Filipe |
| Assunto: | Data mining KDD Real time Pervasive IDSS Intensive care Intelligent decision support system |
| Ano: | 2012 |
| País: | Portugal |
| Tipo de documento: | comunicação em conferência |
| Tipo de acesso: | acesso aberto |
| Instituição associada: | Universidade do Minho |
| Idioma: | inglês |
| Origem: | RepositóriUM - Universidade do Minho |
| Resumo: | The introduction of an Intelligent Decision Support System (IDSS) in a critical area like the Intensive Medicine is a complex and difficult process. In this area, their professionals don’t have much time to document the cases, because the patient direct care is always first. With the objective to reduce significantly the manual records and, enabling, at the same time, the possibility of developing an IDSS which can help in the decision making process, all data acquisition process and knowledge discovery in database phases were automated. From the data acquisition to the knowledge discovering, the entire process is autonomous and executed in real-time. On-line induced data mining models were used to predict organ failure and outcome. Preliminary results obtained with a limited population of patients showed that this approach can be applied successfully. |
|---|