English

Abstract

The analog signals are changed to digital signals for an instance of time designed as a powerful imaging system. The imaging system produced in the digital form as it evolves the analog imaging devices has the ability to perform digital technology. Therefore, the image expansion is guided by the diagnostic system and surgical systems. The present research work used a hybrid compressive sensing algorithm consisting of an optimized neural network and lossy-lossless compression using Adaptively learned sparsifying based on L1 minimization. The Region of Interest (ROI) is compressed using Integer-based Lifting Wavelet Transform with lossless compression, and the non-ROI using an Optimized neural network. The lossy models are irreversible and achieve a higher compression ratio; therefore, the medical image processing has the visual quality in reconstruction showing compression ratio at the highest. The proposed method overcomes the dimensional reduction problem for optimizing with sparsity to non-ROI regions. Therefore, the medical image is transmitted over the network with limited bandwidth. The proposed work outcomes showed that the developed model attained a PSNR of 35.62 dB and SSIM of 0.984 better when compared to the existing ROI-CS Net model, which obtained a PSNR of 29.55 dB and SSIM of 0.89.