The Application Of Improved Genetic Algorithm In Conductivity Imaging Of Stratified Biological Tissue

Electromagnetic measurement is an important way for diagnostic purposes in biomedical applications, and the research of which is a very challenging topic. Conductivity reconstruction, one kind of electromagnetic measurement, reconstructs conductive property distribution of the biological tissue exposed to electromagnetic fields by measuring the fields around it. This thesis presents the application of improved Genetic algorithm to reconstruct the conductivity distribution of stratified biological tissue by measuring the change of impedance in the coil.At first, The thesis briefly introduce the development of electromagnetic measurement, meanwhile some classic methods about it were describedIn the second section, calculating the forward problem, for which will be used for iteration by Genetic Algorithm (GA). The thesis analyzes the forward problem at different circumstances.In inverse calculation, the Micro-GA can be employed. Then the time penalty can be largely diminished during all the searching procedure. However,, the Micro-GA just reaches the near-optimal region, not the precise solution. So the improvement, such as the self-adaptive fitness function combined with the penalty function methods, self-adaptive crossover probability and the BP operator enlightened by neutral network, especially the BP network to improve the local optimal capacity were used.Comparing the SGA and the improved M-GA, the experimental result shows that the improved M-GA largely enhances the performance and presents the higher noise-resistance.