Descripción
|
|
---|---|
Traditionally, clinical data have been the only source of information for disease diagnosis. Today, there are other types of information such as DNA microarrays, which are taken into account to improve diagnosis and prognosis of many diseases. This thesis proposes a new approach, called CliDaPa, to efficiently combine both sources of information (clinical and genetic data), in order to further improve estimations. In this approach, patients are firstly segmented using a tree representation through their clinical data (clinical tree). Therefore, different groups of patients are identified according to similar behavior. Then each individual group is studied with data mining techniques, using the genetic information. To demonstrate its validity, the method is applied to different real data sets (breast and brain cancer). The validation of the results is based on two methods of validation, internal and external, using the Supercomputing and Visualization Centre of Madrid (CeSViMa), where the three approaches of the algorithm were implemented in parallel. The results are compared with other literature studies, as well as traditional analysis techniques, demonstrating a significant improvement over existing results. | |
Internacional
|
No |
ISBN
|
|
Tipo de Tesis
|
Doctoral |
Calificación
|
Sobresaliente cum laude |
Fecha
|
18/10/2010 |