Deep Gupta

Nonsubsampled shearlet based CT and MR medical image fusion using biologically inspired spiking neural network

Abstract This paper presents a new fusion scheme for the CT and MR medical images that utilizes both the features of the nonsubsampled shearlet transform (NSST) and spiking neural network. As a new image representation with the different features, the NSST is utilized to provide an effective representation of the image coefficients. Firstly, the source CT and MR images are decomposed by the NSST into several subimages. The regional energy is used to fuse the low frequency coefficients. High frequency coefficients are also fused using a pulse coupled neural network model that is used as a biologically inspired type neural network. It also retains the edges and detail information from the source images. Finally, the inverse NSST is used to produce the fused image. The performance of the proposed fusion method is evaluated by conducting several experiments on the different CT and MR medical image datasets. Experimental results demonstrate that the proposed method does not only produce better results by successfully fusing the different CT and MR images, but also ensures an improvement in the various quantitative parameters as compared to other existing methods.

A hybrid segmentation method based on Gaussian kernel fuzzy clustering and region based active contour model for ultrasound medical images

Abstract Segmentation is a very crucial task for the ultrasound medical images due to the presence of various imaging artifacts and noise. This paper presents a hybrid segmentation method for the ultrasound medical images that utilize both the features of the Gaussian kernel induced fuzzy C-means (GKFCM) clustering and active contour model driven by region scalable fitting (RSF) energy function. In this method, the result obtained from the GKFCM method is utilized to initialize the contour that spreads to identify the estimated regions. It also helps to estimate the several controlling parameters used in the curve evolution process. The RSF formulation that is responsible for attracting the contour toward the object boundaries removes the requirement of the re-initialization process. The performance of the proposed method is evaluated by conducting several experiments on both the synthetic and real ultrasound images. Experimental results demonstrate that the proposed method produces better results by successfully detecting the object boundaries and also ensures an improvement in segmentation accuracy compared to others.

Despeckling of ultrasound medical images using nonlinear adaptive anisotropic diffusion in nonsubsampled shearlet domain

Abstract Despeckling is of great interest in ultrasound medical images. The inherent limitations of acquisition techniques and systems introduce the speckles in ultrasound images. These speckles are the main factors that degrade the quality and most importantly texture information present in ultrasound images. Due to these speckles, experts may not be able to extract correct and useful information from the images. This paper presents an edge preserved despeckling approach that combines the nonsubsampled shearlet transform (NSST) with improved nonlinear diffusion equations. As a new image representation method with the different features of localization, directionality and multiscale, the NSST is utilized to provide the effective representation of the image coefficients. The anisotropic diffusion approach is applied to the noisy coarser NSST coefficients to improve the noise reduction efficiency and effectively preserves the edge features. In the diffusion process, an adaptive gray variance is also incorporated with the gradient information of eight connected neighboring pixels to preserve the edges, effectively. The performance of the proposed method is evaluated by conducting extensive simulations using both the standard test images and several ultrasound medical images. Experiments show that the proposed method provides an improvement not only in noise reduction but also in the preservation of more edges as compared to several existing methods.

Ripplet domain non-linear filtering for speckle reduction in ultrasound medical images

Abstract Ultrasound imaging is one of the most important and cheapest instrument used for diagnostic purpose among the clinicians. Due to inherent limitations of acquisition methods and systems, ultrasound images are corrupted by the multiplicative speckle noise that degrades the quality and most importantly texture information present in the ultrasound image. In this paper, we proposed an algorithm based on a new multiscale geometric representation as discrete ripplet transform and non-linear bilateral filter in order to reduce the speckle noise in ultrasound images. Ripplet transform with their different features of anisotropy, localization, directionality and multiscale is employed to provide effective representation of the noisy coefficients of log transformed ultrasound images. Bilateral filter is applied to the approximation ripplet coefficients to improve the denoising efficiency and preserve the edge features effectively. The performance of the proposed method is evaluated by conductive extensive simulations using both synthetic speckled and real ultrasound images. Experiments show that the proposed method provides better results of removing the speckle and preserving the edges and image details as compared to several existing methods.

Speckle filtering of ultrasound images using a modified non-linear diffusion model in non-subsampled shearlet domain

Speckle filtering is of great interest for the ultrasound medical images in which various noises and artefacts are introduced because of various limitations of the acquisition systems and techniques. Speckle is a prime factor to degrade the quality and most importantly, texture information present in the ultrasound images. This study presents a despeckling method based on a modified non-linear diffusion model and non-subsampled shearlet transform (NSST). As a new image representation method with the different features of localisation, directionality and multiscale, the NSST is utilised to provide the effective representation of the image coefficients. The modified anisotropic diffusion is applied to the noisy coarser NSST coefficients to improve the denoising efficiency and preserve the edge features effectively. In the diffusion process, the non-local pixel information is incorporated to evaluate the gradient of eight connected neighbouring pixels with an adaptive grey variance. The performance of the proposed method is evaluated for both the standard test and real ultrasound images. Experimental results show that the proposed method produces better results of noise suppression with the preservation of more edges compared with several existing methods.

Despeckling of ultrasound medical images using ripplet domain nonlinear filtering

Despeckling is of great interest for the ultrasound medical images in which various types of noise and artifacts are introduced because of inherent limitations of the acquisition techniques and systems. Among these noise and artifacts, speckle is a main factor, which degrades the quality and most importantly texture information present in the ultrasound images. Due to the speckle, experts may not be able to extract correct and useful information from the images. This paper presents a despeckling method based on a new multiscale geometric representation such as discrete ripplet transform (DRT) and nonlinear bilateral filter (NLBF). The DRT, a new image representation approach with the different features of anisotropy, localization, directionality, and multiscale, is employed to provide effective representation of the noisy coefficients. Bilateral filter is applied to the noisy ripplet coefficient to improve the denoising efficiency and preserve the edge features effectively. The proposed method also helps to improve the visual quality of the ultrasound images. The performance of the proposed method is evaluated on the different ultrasound medical images and results show significant improvement not only in the speckle reduction but also in the edge preservation performance.

Enhancement of Medical Ultrasound Images Using Multiscale Discrete Shearlet Transform Based Thresholding

Feature preserved enhancement is of great interest in medical ultrasound images. Speckle is a main factor which affects the quality, contrast resolution and most importantly texture information present in ultrasound images and can make the post-processing difficult. This paper presents a new enhancement approach which is based on discrete shearlet transform (DST) and thresholding scheme. The DST, a new efficient multiscale geometric representation with the different features of anisotropy, localization, directionality and multiscale, is employed to provide effective representation of the noisy coefficients. Thresholding schemes are applied to the noisy DST coefficients to improve the denoising efficiency and preserve the edge features effectively with this consideration that blurring associated with speckle reduction should be less and fine details are enhanced/preserved properly for the visual enhancement of ultrasound images. The presented algorithm also helps to improve the visual quality of the ultrasound images. Experimental results demonstrate the ability of proposed method for noise suppression, feature and edge preservation defined in terms of different performance measures.

Natural frequencies of a non homogeneous isotropic elastic infinite plate of variable thickness resting on elastic foundation

SommarioSi sono studiate le oscillazioni libere di una piastra elastica infinita, non omogenea, isotropa, di spessore parabolicamente variabile, poggiata su un suolo elastico. Applicando il metodo di Frobenius per la soluzione della equazione differenziale del moto si sono calcolate le frequenze, le deformate ed i momenti corrispondenti ai primi cinque modi di vibrazione per due combinazioni di condizioni al contorno, incastro-incastro ed incastro-appoggio e diversi valori della rastremazione, del parametro di non omogeneità e del modulo del suolo.SummaryThe dynamic free response of a nonhomogeneous isotropic elastic infinite plate of parabolically varying thickness resting on an elastic foundation has been studied. The frequencies, deflections and moments corresponding to the first five modes of vibration have been computed for the two combinations of boundary conditions, clamped-clamped (C-C) and clamped-simply supported (C-SS) and various values of taper constant, nonhomogeneity parameter and foundation modulus by applying the method of Frobenius for the solution of the governing differential equation of motion.

Edge preserved enhancement of medical images using adaptive fusion–based denoising by shearlet transform and total variation algorithm

Abstract. Edge preserved enhancement is of great interest in medical images. Noise present in medical images affects the quality, contrast resolution, and most importantly, texture information and can make post-processing difficult also. An enhancement approach using an adaptive fusion algorithm is proposed which utilizes the features of shearlet transform (ST) and total variation (TV) approach. In the proposed method, three different denoised images processed with TV method, shearlet denoising, and edge information recovered from the remnant of the TV method and processed with the ST are fused adaptively. The result of enhanced images processed with the proposed method helps to improve the visibility and detectability of medical images. For the proposed method, different weights are evaluated from the different variance maps of individual denoised image and the edge extracted information from the remnant of the TV approach. The performance of the proposed method is evaluated by conducting various experiments on both the standard images and different medical images such as computed tomography, magnetic resonance, and ultrasound. Experiments show that the proposed method provides an improvement not only in noise reduction but also in the preservation of more edges and image details as compared to the others.

CT and MR image information fusion scheme using a cascaded framework in ripplet and nonsubsampled shearlet domain

The fusion of multimodal medical information is considered as an assisted approach for the medical professionals. Computed tomography and magnetic resonance (CT–MR) medical image fusion are able to help the radiologist in precise diagnosis of disease and deciding the required treatment in accord with the patients condition. Therefore, a cascaded framework is proposed in this study that presents a fusion approach for multimodal medical information in ripplet transform (RT) and non-subsampled shearlet (NSST) domain. The RT and NSST having different features are utilised in a cascade manner that provides several directional decomposition coefficients and increases shift invariance information in the fused images. At the first stage decomposition, a biologically inspired neural model, motivated by novel sum-modified Laplacian and spatial frequency is utilised to fuse the low- and high-frequency coefficients, respectively, and the max fusion rule based on regional energy is utilised at stage 2. This model is used to preserve the redundant information also. The fusion performance is also validated by extensive simulations performed on different CT–MR image datasets of different diseases. Experimental results demonstrate that the proposed method provides better fused images in terms of visual quality along with the quantitative indices compared with several existing fusion approaches.