         MOHAMMADREZA NASIRIAVANAKI 1 , S. A. HOJJATOLESLAMI 1 , MARIA PAUN 2 , SIMON TUOHY 2 , ALEXANDER MEADWAY 2 , GEORGE DOBRE 2 and ADRIAN GH. PODOLEANU 2 1. Research and Development Centre, School of Biosciences, University of Kent, Canterbury, Kent, CT2 7PD, UNITED KINGDOM 2. Applied Optics Group (AOG), School of Physical Sciences, University of Kent, Canterbury, CT2 7NH, UNITED KINGDOM mn96@kent.ac.uk http://www.kent.ac.uk/physicalsciences/research/aog/   In this study optimized compensation of wavefront aberrations utilizing an electromagnetic deformable mirror is presented. The mirror is controlled by PCI cards and sound card through simulated annealing algorithm implemented by using the integration of Visual C++ and MATLAB in MATLAB environment.   optimization, simulated annealing algorithm, adaptive optics, Matlab   Optical coherence tomography (OCT) is an advanced imaging tool, which produces high resolution threedimensional (3D) images [1]. OCT is a noninvasive imaging modality that applies non ionizing safe optical radiation [1]. OCT images are constructed by measuring the magnitude of the optical echoes at different transverse positions on the sample [2]. The optical devices and the surface of the sample deteriorates the wavefront and produce aberrations. Wavefront aberrations reduce the signal to noise ratio and consequently diminish the contrast of the OCT images. In addition, aberrations reduce the resolution of the imaging system. The aberrations are different from sample to sample, hence it is difficult to correct them by a static technique. Adaptive optics (AO) has been introduced to reduce the effect of the aberrations. It becomes essential especially in the applications demanding a higher image signaltonoise ratio and higher resolution. An AO system is made up of a wavefront sensor, such as a ShackHartmann, to measure the wavefront error, a deformable mirror (DM), to compensate the distortion, and a control loop method, to find the optimized compensation. Unfortunately such systems are expensive [3]. On the other hand, although openloop wavefront or beamshape control can in principle be achieved using the knowledge of a deformable mirror’s influence matrix, its nonlinear characteristics make implementation of a wide range of predeformed shapes problematic without the use of feedback through a wavefront sensor or other forms of closed loop control system. Less expensive closedloop methods are required for correcting distortions, such as the socalled blind optimization or sensorless approaches [47]. In blind optimization algorithms, only a DM is required. Such algorithms operate in a closedloop iteratively to optimize a single measurable variable by reducing the wavefront error obtained by changing the mirror surface shape. Simulated annealing (SA) introduced in 1982 by Kirkpatrick et al. [9], is a technique for combinatorial optimization problems, such as minimizing multivariate functions [1011] to improve the solution to the socalled Travelling Salesman Problem [12]. SA is favoured due to its efficiency with less computational complexity since all other known techniques for obtaining an optimum solution which require an exponentially increasing number of steps as the problems become larger [10]. In comparison with iterative improvement methods [9], which are captured simply in local minima, SA improves the result by occasionally searching in directions that lead to worse solutions [13]. SA is however a heuristic optimization technique which does not guarantee in achieving the optimum value [10]. The concept of MATHEMATICAL METHODS AND APPLIED COMPUTING ISSN: 1790-2769 669 ISBN: 978-960-474-124-3