A hybrid filtering technique in medical image denoising: Blending of neural network and fuzzy inference

Recently, image processing plays a vital role in the medical field because most of the diseases are diagnosed by means of medical images. In order to utilize these images for the diagnosing process, it must be a noiseless one. However, most of the images are affected through noises and artifacts caused by the various acquisition techniques and hence an effective technique for denoising is necessary for medical images particularly in Computed Tomography, which is a significant and most general modality in medical imaging. In order to achieve this denoising of CT images, an effective CT image denoising technique is proposed. The proposed technique confiscates the Additive white Gaussian Noise from the CT images and improves the quality of the CT images. The proposed work is comprised of three phases; they are preprocessing, training and testing. In the preprocessing phase, the CT image which is affected by the AWGN noise is transformed using multi wavelet transformation. In the training phase the obtained multi-wavelet coefficients are given as input to the Adaptive Neuro-Fuzzy Inference System (ANFIS). In the testing phase, the input CT image is examined using this trained ANFIS and then to enhance the quality of the CT image thresholding is applied and then the image is reconstructed. Hence, the denoised and the quality enhanced CT images are obtained in an effective manner.