Medical Image Denoising Expending A Hybrid Recurrent Neural Network Through Support Vector Machine

Medical imaging helps doctors make more informed treatment decisions by providing a visual representation of the anatomy that can't be gleaned from a physical exam alone. Medical diagnosis relies heavily on the visual. Over the past few years, medical imaging has established itself as a crucial tool in the diagnosis of illness. Over the past few of decades, a number of medical imaging modalities have emerged, each with its own unique set of capabilities. These methods allow for the acquisition of images of the internal anatomical structures that need to be inspected, without the need to physically open the body. In this study, a system was suggested that uses a Hybrid Recurrent Neural Networks with Support Vector Machine (HRNN-SVM) method to effectively remove white noise, salt and pepper noise, Gaussian and speckle noise from CT lung images. To enhance RNNs' discrimination skills, the SVM is implemented in the final RNN layer in the HRNN-SVM technique. Batch normalization is combined with residual learning to speed up the learning process and increase accuracy. In this study, LSTM is used to carry out batch normalization. The Firefly Algorithm is used to determine the best batch size for batch normalization (FA). When compared to the PSO algorithm, it reduces the time required for training. To do this, the RNN's hidden layers are gradually splitting apart the image's structure from the noisy observation.