|
This research project aimed to the development of a novel medical decision support
system for cancer diagnosis, using clinical and DNA patient data. The
proposed system consists of two autonomous subsystems. Their outputs may
be combined for the final diagnosis. The first subsystem uses endoscopic
video as input. It provides information regarding the existence and the
location of lesions. The research was focusing to the development of novel
methods for video analysis, real-time feature extraction, selection and
classification. The second subsystem uses DNA microarray data as input.
Microarray data processing determine the genes of which the expression is
differentiated in the presence of cancer. The research was focusing in
data preprocessing, feature selection and cancer identification and/or
prediction methods. Time consuming data processing algorithms have been
implemented in Field Programmable Gate Arrays (FPGA), to achieve realtime
performance. Medical and biological evaluation of the system’s
results have been realized. The results of this research could be proved
valuable for disease diagnosis and prediction and could be used for the
determination of new pharmaceutical targets in the future. |
