Abstract
Cartesian Genetic Programming (CGP) is a form of genetic programming that is flexible and adaptable to a range of problems. In this article, a particular representation of CGP, known as implicit context representation CGP is presented and its application to two medical conditions: the diagnosis of Parkinson' disease and the detection of breast cancer from mammograms. CGP has a number of advantages over conventional genetic programming and is well suited to the highly non-linear problems considered here. Summary results are presented for the application of CGP to real patient data that are sufficiently encouraging to warrant further clinical trials which are currently in progress.
| Original language | English |
|---|---|
| Article number | 6052376 |
| Pages (from-to) | 56-67 |
| Number of pages | 12 |
| Journal | Ieee computational intelligence magazine |
| Volume | 6 |
| Issue number | 4 |
| DOIs | |
| Publication status | Published - Nov 2011 |
Keywords
- PARKINSONS-DISEASE
- IMAGE SEGMENTATION
- IMPLICIT CONTEXT
- MICROCALCIFICATIONS
- CLASSIFICATION
- MAMMOGRAMS
- ALGORITHMS
- FEATURES
- SYSTEM