Wavelet decomposition for speckle reduction with feature preservation in Optical Coherence Tomography

Optical Coherence Tomography (OCT) images exhibit the effects of speckle which can make image interpretation and quantitative measurements difficult. Many approaches have been developed to reduce this speckle including both hardware implementations and post-processing techniques. However, they either suffer from a loss in resolution and blurring of the image or an increase in complexity and reduction in speed of the system. Wavelet decomposition has been shown to effectively separate the resolvable features of an image from the speckle pattern. The two components can then be processed separately. The speckle pattern can be filtered and then recombined with the resolvable component to create an image with improved signal-to-noise-ratio (SNR) and intact image details. The results of this algorithm are demonstrated on in vivo OCT images of skin taken with a swept-source based system. Such a technique, when applied, for example, to OCT images of disease, can be extremely useful in improving the clinical interpretation of the images as well as allowing more accurate quantitative measurements not affected by the presence of speckle.