Speckle noise reduction of optical coherence tomography images with a wavelet transform

摘要

Optical coherence tomography (OCT) is a technique to investigate the microstructure of biological tissue non-invasively. OCT uses a low-coherence interferometry technique, and so the quality of OCT images is degraded by speckle noise which is inherent in coherent imaging. To remove the speckle noise, many methods have been proposed including hardware- and software-based methods. The latter approaches are flexible and easy to implement, and so many methods such as median filter, Wiener filter, and wavelet filter have been proposed. Of these, use of a wavelet filter is a powerful method to reduce the speckle noise, and this filter is often used in ultrasound images. The model of the speckle noise has a multiplicative nature, and conventional filtering methods are somewhat ineffective against this speckle noise. In 2003, we have developed a noise reduction method using a wavelet transform for radiographic images. In this paper, we applied the method to OCT images. The results of simulations showed that the speckle noise was reasonably reduced. And if we had only four frame images, the signal to noise ratio (S/N) was equivalent to that of averaged image using 32 frames. In addition, when we increased the number of processed images, we achieved the S/N of 24-25 dB.