Search

Abstract Accurate and unbiased classification of breast lesions is pivotal for early diagnosis and treatment, and a deep learning approach can effectively represent and utilize the digital content of images for more precise medical image analysis. Breast ultrasound imaging is useful for detecting and distinguishing benign masses from malignant masses. Based on the different ways in which benign and malignant tumors affect neighboring tissues, i.e., the pattern of growth and border irregularities, the penetration degree of the adjacent tissue, and tissue-level changes, we investigated the relationship between breast cancer imaging features and the roles of inter- and extra-lesional tissues and their impact on refining the performance of deep learning classification. The novelty of the proposed approach lies in considering the features extracted from the tissue inside the tumor (by performing an erosion operation) and from the lesion and surrounding tissue (by performing a dilation operation) for classification. This study uses these new features and three pre-trained deep neuronal networks to address the challenge of breast lesion classification in ultrasound images. To improve the classification accuracy and interpretability of the model, the proposed model leverages transfer learning to accelerate the training process. Three modern pre-trained CNN architectures (MobileNetV2, VGG16, and EfficientNetB7) are used for transfer learning and fine-tuning for optimization. There are concerns related to the neuronal networks producing erroneous outputs in the presence of noisy images, variations in input data, or adversarial attacks; thus, the proposed system uses the BUS-BRA database (two classes/benign and malignant) for training and testing and the unseen BUSI database (two classes/benign and malignant) for testing. Extensive experiments have recorded accuracy and AUC as performance parameters. The results indicate that the proposed system outperforms the existing breast cancer detection algorithms reported in the literature. AUC values of 1.00 are calculated for VGG16 and EfficientNet-B7 in the dilation cases. The proposed approach will facilitate this challenging and time-consuming classification task.

Abstract Detecting detonators is a challenging task because they can be easily mis-classified as being a harmless organic mass, especially in high baggage throughput scenarios. Of particular interest is the focus on automated security X-ray analysis for detonators detection. The complex security scenarios require increasingly advanced combinations of computer-assisted vision. We propose an extensive set of experiments to evaluate the ability of Convolutional Neural Network (CNN) models to detect detonators, when the quality of the input images has been altered through manipulation. We leverage recent advances in the field of wavelet transforms and established CNN architectures—as both of these can be used for object detection. Various methods of image manipulation are used and further, the performance of detection is evaluated. Both raw X-ray images and manipulated images with the Contrast Limited Adaptive Histogram Equalization (CLAHE), wavelet transform-based methods and the mixed CLAHE RGB-wavelet method were analyzed. The results showed that a significant number of operations, such as: edges enhancements, altered color information or different frequency components provided by wavelet transforms, can be used to differentiate between almost similar features. It was found that the wavelet-based CNN achieved the higher detection performance. Overall, this performance illustrates the potential for a combined use of the manipulation methods and deep CNNs for airport security applications.

Abstract This paper introduces superpixels to enhance the detection of skin lesions and to discriminate between melanoma and nevi without false negatives, in dermoscopy images. An improved Simple Linear Iterative Clustering (iSLIC) superpixels algorithm for image segmentation in digital image processing is proposed. The local graph cut method to identify the region of interest (i.e., either the nevi or melanoma lesions) has been adopted. The iSLIC algorithm is then exploited to segment sSPs. iSLIC discards all the SPs belonging to image background based on assigned labels and preserves the segmented skin lesions. A shape and geometric feature extraction task is performed for each segmented SP. The extracted features are fed into six machine learning algorithms such as: random forest, support vector machines, AdaBoost, k-nearest neighbor, decision trees (DT), Gaussian Naïve Bayes and three neural networks. These include Pattern recognition neural network, Feed forward neural network, and 1D Convolutional Neural Network for classification. The method is evaluated on the 7-Point MED-NODE and PAD-UFES-20 datasets and the results have been compared to the state-of-art findings. Extensive experiments show that the proposed method outperforms the compared existing methods in terms of accuracy.

By investigating the concatenation model that incorporates a power-law nonlinearity, this paper provides an in-depth analysis, and the soliton solutions are derived. By utilizing the extended tanh-function technique and the enhanced Kudryashov’s scheme, bright and singular 1-solitons are yielded, as demonstrated in this paper. The conservation laws can be obtained through the multipliers approach, which subsequently exposes the conserved quantities.

Addressing complex interactions within water, energy, and food (WEF) resources, innovative tools for in-depth analysis and decision-making are imperative. This study introduces chorematic focus maps (CFMs) as a groundbreaking method to visualize and tackle the WEF nexus’s complexities, focusing specifically on the Danube Delta Biosphere Reserve (DDBR). By merging geospatial analysis with on-site validation, this research reveals intricate interdependencies within the nexus and positions CFMs as an effective tool for stakeholders. This study adopts a methodological approach that focuses on identifying human activities and evaluating their impacts on the WEF nexus, with the goal of developing practical and grounded strategies for managing these essential resources. By testing this approach within the DDBR, the potential for wider application is demonstrated, offering a promising framework for addressing similar socio-environmental challenges across various regions. Future research directions include refining CFMs’ precision and practicality through extended fieldwork and stakeholder engagement, testing the framework’s adaptability across various locations and nexus dynamics. Additionally, incorporating cutting-edge technologies such as machine learning could provide deeper insights and reinforce CFMs’ role in decision support for the WEF nexus. Conclusively, this investigation into the WEF nexus through CFMs emphasizes the critical need for strategies that navigate the complexities of environmental management and resource optimization, marking CFMs as a significant tool for both decision-makers and researchers.

This paper studies the formulation of dark and singular stationary optical solitons that stem from quadratic-cubic and generalized quadratic-cubic forms of nonlinear refractive index coupled with nonlinear chromatic dispersion. The temporal evolution is taken to be of both kinds, namely linear and generalized. The enhanced Kudryashov’s approach enables this retrieval possible.

The primary objective of this research is to investigate and analyze the intricate dynamics exhibited by dispersive optical soliton solutions within dispersion-flattened fibers while considering the impact of white noise. By examining this phenomenon, the study aims to gain a comprehensive understanding of the behavior and characteristics of these solitons, which are essential in the field of optical communication. A wide range of soliton solutions is obtained by the extended auxiliary equation procedure and the unified Riccati equation scheme, which are two powerful methods employed in the field of nonlinear dynamics and soliton theory to obtain a diverse array of soliton solutions. These techniques enable researchers to explore and analyze a wide range of complex nonlinear phenomena in various physical systems. This wide range of soliton solutions contributes to a deeper understanding of nonlinear phenomena and provides valuable insights into the fundamental properties and applications of solitons in different areas of science and engineering, including optics, fluid dynamics, plasma physics, and condensed matter physics. White noise only affects the phase part of the solitons and does not spread to other components. Such conclusions are applicable only under the specific transformation and the special simplification employed in deriving the solutions in the paper. The existence of white noise does not change the soliton amplitudes, widths, or velocities in any component. This paper is the first to report an essential observation concerning dispersion-flattened fibers, which has never been reported before.

The current study focuses on the recovery of quiescent optical solitons through the use of the complex Ginzburg–Landau equation when the chromatic dispersion is rendered to be nonlinear. A dozen forms of self-phase modulation structures are taken into consideration. The utilization of the enhanced Kudryashov's scheme has led to the emergence of singular, dark, and bright soliton solutions. The existence of such solitons is subject to certain parametric restrictions, which are also discussed in this paper.

The present study reports quiescent optical solitons for Kudryashov’s proposed form of self-phase modulation with quintuple nonlinearity having two power-law parameters. The solitons were retrieved with the usage of enhanced Kudryashov’s algorithm. The stationary singular, dark and bright solitons emerged from the scheme together with parametric restrictions. Both linear and generalized temporal evolution terms are considered.

The current paper implements three elegant approaches to recover a complete spectrum of optical solitons to the Radhakrishnan–Kundu–Lakshmanan equation with dual–power law of nonlinear refractive index. The conservation laws are also recovered by the usage of multipliers approach. The parameter constraints for the existence of such solitons are also enumerated. The numerical simulations of the recovered soliton solutions are also presented.

The current paper recovers cubic–quartic optical solitons in fiber Bragg gratings having polynomial law of nonlinear refractive index structures. Lie symmetry analysis is carried out, starting with the basic analysis. Then, it is followed through with improved Kudryashov and generalized Arnous schemes. The parameter constraints are also identified for the existence of such solitons. Numerical surface plots support the adopted applied analysis.

In the current study, the concatenation model of birefringent fibers is explored for the first time, and we present optical soliton solutions to the model. The integration algorithm used to achieve this retrieval is the method of undetermined coefficients, which yields a wide range of soliton solutions. The parameter constraints arise naturally during the derivation of the soliton solutions, which are essential for such solitons to exist.

The paper revisits highly dispersive optical solitons that are addressed by the aid of Lie symmetry followed by the implementation of the Riccati equation approach and the improved modified extended tanh-function approach. The soliton solutions are recovered and classified. The conservation laws are also recovered and the corresponding conserved quantities are enlisted.

This paper recovers quiescent optical solitons that are self-sustaining, localized wave packets that maintain their shape and amplitude over long distances due to a balance between nonlinearity and dispersion. When a soliton is in a state of quiescence, it means that it is stationary in both space and time. Quiescent optical solitons are typically observed in optical fibers, where nonlinearity and dispersion can lead to the formation of solitons. The concatenation model is considered to understand the behavior of optical pulses propagating through nonlinear media. Here, we consider the familiar nonlinear Schrödinger equation, the Lakshmanan–Porsezian–Daniel equation, and the Sasa–Satsuma equation. The current paper also addresses the model with nonlinear chromatic dispersion, a phenomenon that occurs in optical fibers and other dispersive media, where the chromatic dispersion of the material is modified by nonlinear effects. In the presence of nonlinearities, such as self-phase modulation and cross-phase modulation, the chromatic dispersion coefficient becomes a function of the optical intensity, resulting in nonlinear chromatic dispersion. A full spectrum of stationary optical solitons, along with straddled stationary solitons, are obtained. There are four integration schemes that made this retrieval possible. The numerical simulations are also included for these solitons. The parameter constraints also indicate the existence criteria for these quiescent solitons.

Purpose: In this article, two main subjects are discussed. First, the nonlinear Schrödinger equation (NLSE) with an anti-cubic (AC) nonlinearity equation is examined, which has a great working area, importance and popularity among the study areas of soliton behavior in optical fibers, by using the enhanced modified extended tanh expansion method and a wide range of optical soliton solutions is obtained. Second, the effects of AC parameters on soliton behavior are examined for each obtained soliton type. Methodology: In order to apply the method, the non-linear ordinary differential equation form (NLODE) of the investigated NLSE-AC is obtained by applying the defined wave transformation. Then, with the help of the proposed algorithm for the NLODE form, polynomial form, an algebraic equation system is obtained by setting the coefficients of this form to zero, and the solution of this system is also obtained. After determining the suitable solution set, the optical soliton solution of the investigated problem is obtained with the help of the serial form of the proposed method, a Riccati solution and wave transform. After checking that the solution satisfies the investigated problem, 3D and 2D graphics are obtained for the special parameter values and the necessary comments are made in the relevant sections. Findings: With the proposed method, many optical soliton solutions, such as topological, anti-peaked, combined peaked-bright, combined anti-peaked dark, singular, combined singular-anti peaked, periodic singular, composite kink anti-peaked, kink, periodic and periodic, with different amplitudes are obtained, and 3D and 2D representations have been made. Then, the effect of AC parameters on the soliton behavior in each case has been successfully studied. It has been shown that AC parameters have a significant effect on the soliton behavior, and this effect changes depending on the soliton shape and the parameters. Moreover, providing and maintaining the delicate balance between the soliton shape and the parameters and the interaction of the parameters with each other involves great difficulties. Originality: Although some soliton types of the NLSE-AC equation have been presented for the first time in this study, there is no study in the literature showing the effect of AC parameters on soliton behavior, especially for the NLSE-AC equation.

This paper recovers a broad spectrum of optical solitons for the perturbed nonlinear Schrödinger’s equation having a dual-power law of nonlinearity. The perturbation terms are from inter-modal dispersion and self-frequency shift. The integration scheme is the improved extended tanh function approach. The parameter constraints that naturally emerge are also enumerated.

The paper concentrates on the solitary waves that are retrievable from the generalized Boussinesq equation. The numerical simulations are displayed in the paper that gives a visual perspective to the model studied in neurosciences. The Laplace–Adomian decomposition scheme makes this visualization of the solitons possible. The numerical simulations are being reported for the first time using an elegant approach. The results would be helpful for neuroscientists and clinical studies in Medicine. The novelty lies in the modeling that is successfully conducted with an impressively small error measure. In the past, the model was integrated analytically only to recover soliton solutions and its conserved quantities.

This article studies highly dispersive optical solitons without of self-phase modulation effect. The numerical algorithm implemented in this work is Laplace-Adomian decomposition method. Both bright and dark solitons are addressed. The error measure for the adopted scheme is impressively low.

This paper implements the enhanced Kudryashov approach to retrieve highly dispersive optical solitons and study it with four nonlinear forms. These are the power law, generalized quadratic-cubic law, triple-power law, and the generalized non-local law. This approach reveals bright and singular optical solitons along with the respective parameter constraints.

This paper retrieves an optical 1–soliton solution to a model that is written as a concatenation of the Lakshmanan–Porsezian–Daniel model and Sasa–Satsuma equation. The method of undetermined coefficients obtains a full spectrum of 1–soliton solutions. The multiplier approach yields the conserved densities, which subsequently lead to the conserved quantities from the bright 1–soliton solution.

This paper studies dispersive bright and dark optical solitons, modeled by the Schrödinger–Hirota equation, numerically by the aid of the Adomian decomposition. The surface plots of the algorithm yielded an impressively small measure. The effects of soliton radiation are ignored.

This paper implements the trial equation approach to retrieve cubic–quartic optical solitons in fiber Bragg gratings with the aid of the trial equation methodology. Five forms of nonlinear refractive index structures are considered. They are the Kerr law, the parabolic law, the polynomial law, the quadratic–cubic law, and the parabolic nonlocal law. Dark and singular soliton solutions are recovered along with Jacobi’s elliptic functions with an appropriate modulus of ellipticity.

The current paper considers the enhanced Kudryashov’s technique to retrieve solitons with a governing model having cubic-quintic-septic-nonic and quadrupled structures of self-phase modulation. The results prove that it is redundant to extend the self-phase modulation beyond cubic-quintic nonlinearity or dual-power law of nonlinearity.

This work recovers cubic-quartic optical solitons with dispersive reflectivity in fiber Bragg gratings and parabolic law of nonlinearity. The Lie symmetry analysis first reduces the governing partial differential equations to the corresponding ordinary differential equations which are subsequently integrated. This integration is conducted using two approaches which are the modified Kudryashov’s approach as well as the generalized Arnous’ scheme. These collectively yielded a full spectrum of cubic-quartic optical solitons that have been proposed to control the depletion of the much-needed chromatic dispersion. They range from bright, dark, singular to combo solitons. These solitons are considered with dispersive reflectivity, which maintains the necessary balance between chromatic dispersion and nonlinear refractive index structure for an uninterrupted transmission of solitons along intercontinental distances. Their respective surface and contour plots are also exhibited. A few closing words are included with some prospective future avenues of research to extend this topic further.

The paper derives stationary optical solitons with nonlinear chromatic dispersion. A nonlocal form of nonlinearity and quintuple power–law of nonlinearity are considered. The Kudryashov’s integration scheme enables to retrieve such solitons. A plethora of solitons come with this algorithm.

The current work addresses cubic–quartic solitons to compensate for the low count of the chromatic dispersion that is one of the major hindrances of soliton transmission through optical fibers. Thus, the present paper handles the cubic–quartic version of the perturbed Fokas–Lenells equation that governs soliton communications across trans-oceanic and trans-continental distances. The model is also considered with the power-law form of nonlinear refractive index that is a sequel to the previously reported result. This is a tremendous advancement to the previously known result that was only with the Kerr-law form of nonlinear refractive index. The present paper mainly contributes by generalizing the Kerr law of nonlinearity to the power law of nonlinearity. The prior results therefore fall back as a special case to the results of this paper. The semi-inverse variational principle yields a bright 1-soliton solution that is imperative for the telecommunication engineers to carry out experimental investigation before the rubber meets the road. Hamiltonian perturbation terms are included that come with maximum intensity. The soliton amplitude–width relation is retrievable from a polynomial equation with arbitrary degree. The parameter constraints are also identified for the soliton to exist.

In this paper, the optical solitons for the complex Ginzburg–Landau equation with Kudryashov’s law of refractive index are established. An improved modified extended tanh–function technique is used to extract numerous solutions. Bright and dark solitons, as well as singular soliton solutions, are achieved. In addition, as the modulus of ellipticity approaches unity or zero, solutions are formulated in terms of Jacobi’s elliptic functions, which provide solitons and periodic wave solutions.

A Gaussian mixture model (GMM)-based classification technique is employed for a quantitative global assessment of brain tissue changes by using pixel intensities and contrast generated by b-values in diffusion tensor imaging (DTI). A hemisphere approach is also proposed. A GMM identifies the variability in the main brain tissues at a macroscopic scale rather than searching for tumours or affected areas. The asymmetries of the mixture distributions between the hemispheres could be used as a sensitive, faster tool for early diagnosis. The k-means algorithm optimizes the parameters of the mixture distributions and ensures that the global maxima of the likelihood functions are determined. This method has been illustrated using 18 sub-classes of DTI data grouped into six levels of diffusion weighting (b = 0; 250; 500; 750; 1000 and 1250 s/mm2) and three main brain tissues. These tissues belong to three subjects, i.e., healthy, multiple haemorrhage areas in the left temporal lobe and ischaemic stroke. The mixing probabilities or weights at the class level are estimated based on the sub-class-level mixing probability estimation. Furthermore, weighted Euclidean distance and multiple correlation analysis are applied to analyse the dissimilarity of mixing probabilities between hemispheres and subjects. The silhouette data evaluate the objective quality of the clustering. By using a GMM in the present study, we establish an important variability in the mixing probability associated with white matter and grey matter between the left and right hemispheres. Keywords: Gaussian mixture model, Brain hemispheres, Weight distribution, Weighted Euclidean distance, Clustering, Cluster validity

This paper obtains highly dispersive optical solitons in fiber Bragg gratings with the Kerr law of a nonlinear refractive index. The generalized Kudryashov’s approach as well as its newer version makes this retrieval possible. A full spectrum of solitons is thus recovered.

Differentiating between malignant and benign masses using machine learning in the recognition of breast ultrasound (BUS) images is a technique with good accuracy and precision, which helps doctors make a correct diagnosis. The method proposed in this paper integrates Hu’s moments in the analysis of the breast tumor. The extracted features feed a k-nearest neighbor (k-NN) classifier and a radial basis function neural network (RBFNN) to classify breast tumors into benign and malignant. The raw images and the tumor masks provided as ground-truth images belong to the public digital BUS images database. Certain metrics such as accuracy, sensitivity, precision, and F1-score were used to evaluate the segmentation results and to select Hu’s moments showing the best capacity to discriminate between malignant and benign breast tissues in BUS images. Regarding the selection of Hu’s moments, the k-NN classifier reached 85% accuracy for moment M1 and 80% for moment M5 whilst RBFNN reached an accuracy of 76% for M1. The proposed method might be used to assist the clinical diagnosis of breast cancer identification by providing a good combination between segmentation and Hu’s moments.

This work retrieves polarized optical soliton solutions for pulses in birefringent fibers that are modeled by the Lakshmanan–Porsezian–Daniel model. The unified approach recovers singular solitons only. This approach fails to retrieve the much needed bright and dark soliton solutions. These singular solitons exist with restricted parametric conditions that are also exhibited.

This study presents an algorithm for fingerprint classification using a CNN (convolutional neural network) model and making use of full images belonging to four digital databases. The main challenge that we face in fingerprint classification is dealing with the low quality of fingerprints, which can impede the identification process. To overcome these restrictions, the proposed model consists of the following steps: a preprocessing stage which deals with edge enhancement operations, data resizing, data augmentation, and finally a post-processing stage devoted to classification tasks. Primarily, the fingerprint images are enhanced using Prewitt and Laplacian of Gaussian filters. This investigation used the fingerprint verification competition with four databases (FVC2004, DB1, DB2, DB3, and DB4) which contain 240 real fingerprint images and 80 synthetic fingerprint images. The real images were collected using various sensors. The innovation of the model is in the manner in which the number of epochs is selected, which improves the performance of the classification. The number of epochs is defined as a hyper-parameter which can influence the performance of the deep learning model. The number of epochs was set to 10, 20, 30, and 50 in order to keep the training time at an acceptable value of 1.8 s/epoch, on average. Our results indicate the overfitting of the model starting with the seventh epoch. The accuracy varies from 67.6% to 98.7% for the validation set, and between 70.2% and 75.6% for the test set. The proposed method achieved a very good performance compared to the traditional hand-crafted features despite the fact that it used raw data and it does not perform any handcrafted feature extraction operations.

This paper is a numerical simulation of highly dispersive optical solitons in birefringent fibers with polynomial nonlinear form, which is achieved for the first time. The algorithmic approach is applied with the usage of the Laplace–Adomian decomposition scheme. Dark and bright soliton simulations are presented. The error measure has a very low count, and thus, the simulations are almost an exact replica of such solitons that analytically arise from the governing system. The suggested iterative scheme finds the solution without any discretization, linearization, or restrictive assumptions.

This paper secures solitary waves and conservation laws to the familiar Korteweg–de Vries equation and Gardner’s equation with three dispersion sources. The traveling wave hypothesis leads to the emergence of such waves. The three sources of dispersion are spatial dispersion, spatio–temporal dispersion and the dual-emporal–spatial dispersion. The conservation laws are enumerated for these models, evolved from the multiplier approach. The conserved quantities are computed with the solitary wave solutions that were recovered.

This work analytically recovers the highly dispersive bright 1–soliton solution using for the perturbed complex Ginzburg–Landau equation, which is studied with three forms of nonlinear refractive index structures. They are Kerr law, parabolic law, and polynomial law. The perturbation terms appear with maximum allowable intensity, also known as full nonlinearity. The semi-inverse variational principle makes this retrieval possible. The amplitude–width relation is obtained by solving a cubic polynomial equation using Cardano’s approach. The parameter constraints for the existence of such solitons are also enumerated.

The current manuscript displays elegant numerical results for cubic-quartic optical solitons associated with the perturbed Fokas–Lenells equations. To do so, we devise a generalized iterative method for the model using the improved Adomian decomposition method (ADM) and further seek validation from certain well-known results in the literature. As proven, the proposed scheme is efficient and possess a high level of accuracy.

This paper employs Lie symmetry analysis to recover cubic–quartic optical soliton solutions to the Lakshmanan–Porsezian–Daniel model in birefringent fibers. The results are a sequel to the previously reported work on the same model in unpolarized fibers. Dark, singular, and straddled optical solitons that emerged from the scheme are presented.

In the automatic detection framework, there have been many attempts to develop models for real-time melanoma detection. To effectively discriminate benign and malign skin lesions, this work investigates sixty different architectures of the Feedforward Back Propagation Network (FFBPN), based on shape asymmetry for an optimal structural design that includes both the hidden neuron number and the input data selection. The reason for the choice of shape asymmetry was based on the 5–10% disagreement between dermatologists regarding the efficacy of asymmetry in the diagnosis of malignant melanoma. Asymmetry is quantified based on lesion shape (contour), moment of inertia of the lesion shape and histograms. The FFBPN has a high architecture flexibility, which indicates it as a favorable tool to avoid the over-parameterization of the ANN and, equally, to discard those redundant input datasets that usually result in poor test performance. The FFBPN was tested on four public image datasets containing melanoma, dysplastic nevus and nevus images. Experimental results on multiple benchmark data sets demonstrate that asymmetry A2 is a meaningful feature for skin lesion classification, and FFBPN with 16 neurons in the hidden layer can model the data without compromising prediction accuracy.

A new solution to overcome the constraints of multimodality medical intra-subject image registration is proposed, using the mutual information (MI) of image histogram-oriented gradients as a new matching criterion. We present a rigid, multi-modal image registration algorithm based on linear transformation and oriented gradients for the alignment of T2-weighted (T2w) images (as a fixed reference) and diffusion tensor imaging (DTI) (b-values of 500 and 1250 s/mm2) as floating images of three patients to compensate for the motion during the acquisition process. Diffusion MRI is very sensitive to motion, especially when the intensity and duration of the gradient pulses (characterized by the b-value) increases. The proposed method relies on the whole brain surface and addresses the variability of anatomical features into an image stack. The sparse features refer to corners detected using the Harris corner detector operator, while dense features use all image pixels through the image histogram of oriented gradients (HOG) as a measure of the degree of statistical dependence between a pair of registered images. HOG as a dense feature is focused on the structure and extracts the oriented gradient image in the x and y directions. MI is used as an objective function for the optimization process. The entropy functions and joint entropy function are determined using the HOGs data. To determine the best image transformation, the fiducial registration error (FRE) measure is used. We compare the results against the MI-based intensities results computed using a statistical intensity relationship between corresponding pixels in source and target images. Our approach, which is devoted to the whole brain, shows improved registration accuracy, robustness, and computational cost compared with the registration algorithms, which use anatomical features or regions of interest areas with specific neuroanatomy. Despite the supplementary HOG computation task, the computation time is comparable for MI-based intensities and MI-based HOG methods.

Image registration has an imperative role in medical imaging. In this work, a grey-wolf optimizer (GWO)-based non-rigid demons registration is proposed to support the retinal image registration process. A comparative study of the proposed GWO-based demons registration framework with cuckoo search, firefly algorithm, and particle swarm optimization-based demons registration is conducted. In addition, a comparative analysis of different demons registration methods, such as Wang’s demons, Tang’s demons, and Thirion’s demons which are optimized using the proposed GWO is carried out. The results established the superiority of the GWO-based framework which achieved 0.9977 correlation, and fast processing compared to the use of the other optimization algorithms. Moreover, GWO-based Wang’s demons performed better accuracy compared to the Tang’s demons and Thirion’s demons framework. It also achieved the best less registration error of 8.36 × 10−5.

(1) Background: In this research, we aimed to identify and validate a set of relevant features to distinguish between benign nevi and melanoma lesions. (2) Methods: Two datasets with 70 melanomas and 100 nevi were investigated. The first one contained raw images. The second dataset contained images preprocessed for noise removal and uneven illumination reduction. Further, the images belonging to both datasets were segmented, followed by extracting features considered in terms of form/shape and color such as asymmetry, eccentricity, circularity, asymmetry of color distribution, quadrant asymmetry, fast Fourier transform (FFT) normalization amplitude, and 6th and 7th Hu’s moments. The FFT normalization amplitude is an atypical feature that is computed as a Fourier transform descriptor and focuses on geometric signatures of skin lesions using the frequency domain information. The receiver operating characteristic (ROC) curve and area under the curve (AUC) were employed to ascertain the relevance of the selected features and their capability to differentiate between nevi and melanoma. (3) Results: The ROC curves and AUC were employed for all experiments and selected features. A comparison in terms of the accuracy and AUC was performed, and an evaluation of the performance of the analyzed features was carried out. (4) Conclusions: The asymmetry index and eccentricity, together with F6 Hu’s invariant moment, were fairly competent in providing a good separation between malignant melanoma and benign lesions. Also, the FFT normalization amplitude feature should be exploited due to showing potential in classification.

Background: The purpose of this article is to provide a new evaluation tool based on skeleton maps to assess the tumoral and non-tumoral regions of the 2D MRI in PD-weighted (proton density) and T2w (T2-weighted type) brain images. Methods: The proposed method investigated inter-hemisphere brain tissue similarity using a mask in the right hemisphere and its mirror reflection in the left one. At the hemisphere level and for each ROI (region of interest), a morphological skeleton algorithm was used to efficiently investigate the similarity between hemispheres. Two datasets with 88 T2w and PD images belonging to healthy patients and patients diagnosed with glioma were investigated: D1 contains the original raw images affected by Rician noise and D2 consists of the same images pre-processed for noise removal. Results: The investigation was based on structural similarity assessment by using the Structural Similarity Index (SSIM) and a modified Jaccard metrics. A novel S-Jaccard (Skeleton Jaccard) metric was proposed. Cluster accuracy was estimated based on the Silhouette method (SV). The Silhouette coefficient (SC) indicates the quality of the clustering process for the SSIM and S-Jaccard. To assess the overall classification accuracy an ROC curve implementation was carried out. Conclusions: Consistent results were obtained for healthy patients and for PD images of glioma. We demonstrated that the S-Jaccard metric based on skeletal similarity is an efficient tool for an inter-hemisphere brain similarity evaluation. The accuracy of the proposed skeletonization method was smaller for the original images affected by Rician noise (AUC = 0.883 (T2w) and 0.904 (PD)) but increased for denoised images (AUC = 0.951 (T2w) and 0.969 (PD)).