Wyniki wyszukiwania - Biblioteka Nauki

1

Automatic segmentation of infant brain MR images: With special reference to myelinated white matter

100%

Devi C. N. , Chandrasekharan A. , Sundararaman V. K. , Alex Z. C.

Biocybernetics and Biomedical Engineering

|

2017

|

tom Vol. 37, no. 1

143--158

EN

Automatic segmentation of infant brain images is faced with numerous challenges like poor image contrast, motion artifacts, and changes caused by progressive myelination of the infant brain. Since timely myelination points to normal brain maturity, monitoring the progress and degree of myelination is clinically significant. However, most of the existing segmentation methods do not segment myelinated portions of the infant brain. In this paper, we propose a segmentation approach focused on segmenting the myelinated white matter tissue in T1-weighted magnetic resonance images of the infant brain. The novelty of the algorithm lies in the introduction of a weighted localized Tsallis entropy based thresh-olding method. The proposed method is also tested on older babies beyond the one-year age mark to verify its utility and robustness. It is seen that the mean Dice coefficients obtained for myelin segmentation by the proposed weighted localized method are higher than that of the other methods, namely, the conventional Tsallis entropy thresholding and modified localized method.

2

Optimization driven Deep Convolution Neural Network for brain tumor classification

100%

Kumar S. , Mankame D. P.

Biocybernetics and Biomedical Engineering

|

2020

|

tom Vol. 40, no. 3

1190--1204

EN

The classification and segmentation of the tumor is an interesting area that differentiates the tumorous cells and the non-tumorous cells to identify the tumor level. The segmentation from MRI is a challenge because of its varying sizes of images and huge datasets. Different techniques were developed in the literature for brain tumor classification but due to accuracy and ineffective decision making, the existing techniques failed to provide improved classification. This work introduces an optimized deep learning mechanism; named Dolphin-SCA based Deep CNN, to improve the accuracy and to make effective decisions in classification. Initially, the input MRI images are given to the pre-processing and then, subjected to the segmentation process. The segmentation process is carried out using a fuzzy deformable fusion model with Dolphin Echolocation based Sine Cosine Algorithm (Dolphin-SCA). Then, the feature extraction process is performed based on power LDP and statistical features, like mean, variance, and skewness. The extracted features are used in the Deep Convolution Neural Network (Deep CNN) for performing the brain tumor classification with Dolphin-SCA as the training algorithm. The experimentation is performed using the MRI images taken from the BRATS database and SimBRATS, and the proposed technique has shown superior performance with a maximum accuracy of 0.963.

3

Bayesian HCS-based multi-SVNN: A classification approach for brain tumor segmentation and classification using Bayesian fuzzy clustering

100%

Raju A. R. , Suresh P. , Rao R. R.

Biocybernetics and Biomedical Engineering

|

2018

|

tom Vol. 38, no. 3

646--660

EN

Brain tumor segmentation and classification is the interesting area for differentiating the tumerous and the non-tumerous cells in the brain and to classify the tumerous cells for identifying its level. The conventional methods lack the automatic classification and they consumed huge time and are ineffective in decision-making. To overcome the challenges faced by the conventional methods, this paper proposes the automatic method of classification using the Harmony-Crow Search (HCS) Optimization algorithm to train the multi-SVNN classifier. The brain tumor segmentation is performed using the Bayesian fuzzy clustering approach, whereas the tumor classification is done using the proposed HCS Optimization algorithm-based multi-SVNN classifier. The proposed method of classification determines the level of the brain tumor using the features of the segments generated based on Bayesian fuzzy clustering. The robust features are obtained using the information theoretic measures, scattering transform, and wavelet transform. The experimentation performed using the BRATS database conveys proves the effectiveness of the proposed method and the proposed HCS-based tumor segmentation and classification achieves the classification accuracy of 0.93 and outperforms the existing segmentation methods.

4

Whole kidney and renal cortex segmentation in contrast-enhanced MRI using a joint classification and segmentation convolutional neural network

100%

Klepaczko A. , Majos M. , Stefańczyk L. , Eikefjord E. , Lundervold A.

Biocybernetics and Biomedical Engineering

|

2022

|

tom Vol. 42, no. 1

295--311

EN

Contrast-enhanced magnetic resonance imaging (CE-MRI) is one of the methods routinely used in clinics for the diagnosis of renal impairments. It allows assessment of kidney perfusion and also visualization of various lesions and tissue atrophy due to e.g. renal artery stenosis (RAS). An important indicator of the renal tissue state is the volume and shape of the kidney. Therefore it is highly desirable to equip radiological units in clinics with the software capable of automatic segmentation of the kidneys in CE-MRI images. This paper proposes a solution to this task using an original architecture of a deep neural network. The proposed design employs a three-branch convolutional neural network specialized in: 1) detection of renal parenchyma within an MR image patch, 2) segmentation of the whole kidney and 3) annotation of the renal cortex. We tested our architecture for normal kidneys in healthy subjects and for poorly perfused organs in RAS patients. The accuracy of renal parenchyma segmentation was equal to 0.94 in terms of the intersection over union (IoU) ratio. Accuracy of the cortex segmentation depends on the level of tissue health condition and ranges from 0.76 up to 0.92 of IoU.

5

Automatic contrast enhancement of brain MR images using Average Intensity Replacement based on Adaptive Histogram Equalization (AIR-AHE)

100%

Isa I. S. , Sulaiman S. N. , Mustapha M. , Karim N. K. A.

Biocybernetics and Biomedical Engineering

|

2017

|

tom Vol. 37, no. 1

24--34

EN

Medical imaging is the most established technique of visualizing the interior of the human body without the risk of the non-invasive effect. This technology is designed to produce images, and it is also capable of representing information about the screening location. In MRI imaging, the poor image quality particularly the low contrast image may provide insufficient data for the visual interpretation of such affected locations. Therefore, the need of image enhancement arises to improve image visions and also to computationally support the image processing technique. In general, conventional contrast enhancement methods may work well for some images. However, in MRI brain image, there are often more complex situations where the WMH signal is high but it may mistakenly be considered as other brain tissues such as CSF. With the motivation to classify the most possible WMH regions, this paper proposes a novel image contrast algorithm of WMH enhancement for MRI image. This algorithm is also known as the Average Intensity Replacement – Adaptive Histogram Equalization (AIR-AHE). The proposed algorithm is applied to the FLAIR image based on the intensity adjustment and contrast mapping techniques. The proposed algorithm for the image enhancement is superior to the existing methods by using image evaluation quantitative methods of PSNR, average gradient values and MSE. Furthermore, the edge information pertaining to the potential WMH regions can effectively increase the accuracy of the results.

6

Analysis and comparison of symmetry based lossless and perceptually lossless algorithms for volumetric compression of medical images

80%

Chandrika B. K. , Aparna P. , Sumam D. S.

Journal of Medical Informatics & Technologies

|

2015

|

tom Vol. 24

147--154

EN

Modern medical imaging techniques produce huge volume of data from stack of images generated in a single examination. To compress them several volumetric compression techniques have been proposed. Performance of these compression schemes can be improved further by considering the anatomical symmetry present in medical images and incorporating the characteristics of human visual system. In this paper a volumetric medical image compression algorithm is presented in which perceptual model is integrated with a symmetry based lossless scheme. Symmetry based lossless and perceptually lossless algorithms were evaluated on a set of three dimensional medical images. Experimental results show that symmetry based perceptually lossless coder gives an average of 8.47% improvement in bit per pixel without any perceivable degradation in visual quality against the lossless scheme.