An Intelligent Recurrent Neural Network with Long Short-Term Memory (LSTM) BASED Batch Normalization for Medical Image Denoising

• Published in: Journal of medical systems Vol. 43; no. 8; pp. 234 - 10
• Main Authors: Rajeev, R., Samath, J. Abdul, Karthikeyan, N. K.
• Format: Journal Article
• Language: English
• Published: New York Springer US 01-08-2019; Springer Nature B.V
• Subjects: Algorithms; Health Informatics; Health Sciences; Humans; Image & Signal Processing; Image processing; Image Processing, Computer-Assisted - methods; Information processing; Long short-term memory; Lung - diagnostic imaging; Lungs; Medical imaging; Medicine; Medicine & Public Health; Memory, Short-Term; Neural networks; Neural Networks, Computer; Noise reduction; Particle swarm optimization; Recall; Recurrent neural networks; Scientific papers; Short term; Signal processing; Signal to noise ratio; Statistics for Life Sciences; Tomography, X-Ray Computed; Wearable Computing Techniques for Smart Health; Long short-term memory; Recurrent neural network; Batch normalization; Image denoising
• ISSN: 0148-5598; 1573-689X
• DOI: 10.1007/s10916-019-1371-9

The process of denoising of medical images that are corrupted by noise is considered as a long established setback in the signal or image processing domain. An effective system for denoising in order to remove white, salt and also pepper noises by means of merging the Long Short-Term Memory, otherwise known as LSTM, based Batch Normalization and Recurrent Neural Network or RNN techniques have been proposed in this research paper. The images of the lung CT are considered as an input in this particular work. Following this, an effectual batch size is calculated by employing the method of Particle Swarm Optimization (PSO). To denoise the image, Recurrent Neural Network or RNN algorithm were proposed, here to reduce the internal covariate shift present in the neural networks, the Long Short-Term Memory or LSTM based Batch Normalization is brought-in. With respect to SNR or Peak Signal to Noise Ratio and Mean Square Error (MSE), operations were assessed. This algorithm is considered as competitive to other denoising schemes which have been confirmed by the experimental outcomes.