Search

The process of aligning one image, in the coordinate system of another is called registration. Image registration is vital in the medical domain and is used for diagnosis, therapy planning, and treatment of diseases. Due to its huge applicability and importance, medical image registration has emerged to be a separate domain of research. Beginning from the invasive landmark based registration, innumerable algorithms have been proposed to register two images of the human physiology. However, a breakthrough in the literature occurred when an information theory based measure, mutual information was used to register two images. Since then a large number of new algorithms have developed, which use mutual information for fully automatic registration of medical images. This paper is a survey of these algorithms. Beginning from its development, it discusses about some of the major works done on mutual information based image registration. Some comparative studies with other algorithms have also been discussed. The paper ends with a discussion of the developments in medical image registration algorithms post mutual information, which primarily include deep neural network (DNN) based algorithms.

This work presents an algorithm to polygonize the surface of triangulated 3D digital objects by unifying the edge-adjacent face triangles having similar norms. If the cosine of the dihedral angle between two faces is above a predefined threshold, then the faces are considered to be similar with respect to their norms. By applying the Least Square Method on the boundary vertices of the set of the merged face triangles, a new norm is computed, optimizing the norms of the individual face triangles constituting the merged set. The coordinates of the boundary vertices are recomputed based on this new norm. Final coordinates are the mean value of all the instances of each vertex. As a consequence, many of the vertices of the original input object are removed. In effect, the object can be described by a number of polygons, which is significantly less than the number of original face triangles. Hence, a substantial amount of compression is achieved. Also, the shape and structure of the object can be discerned from the polygons representing it. This process may further be utilised in improved shape analysis. Distortion is computed to illustrate that a trade-off may be made between distortion and compression.

Energy optimization and coverage area optimization of wireless sensor networks (WSN) are two major challenges to accomplish reliability optimization in the field of WSN. Reliability optimization in the field of WSN is directly connected to the performance and efficiency and consistency of the network. In this paper, the authors describe how these challenges can be resolved by designing an efficient WSN with the help of meta-heuristic algorithms. They have configured an optimized route/path using ant colony optimization (ACO) algorithm and deployed static WSN nodes. After configuring an efficient network, if we can maximize the coverage area, then we can ensure that the network is a reliable network. For coverage area optimization, they used a hybrid differential evolution-quantum behaved particle swarm optimization (DE-QPSO) algorithm. The result has been compared with existing literature, and the authors found good results applying those meta-heuristic and hybrid algorithms.

Aim: This paper proposed the design and development of 4-Bit Gray Code Counter Circuit Using Reversible Logic Gate. Methods: The 4-Bit Gray Code Counter Circuit can be design using SAM gate, Feynman gate (FG), double Feynman gate (DFG) and NOT gate. The proposed circuit is the combined application of 4-bit binary asynchronous counter and 4-bit binary to gray code converter circuit. Results: The proposed gray code counter is designed using four no. of SAM gate, three no. of DFG, one FG and seven reversible NOT gate. The QC, GO and CI of proposed circuit are correspondingly 30, 4 and 7. Conclusion: The novel reversible gray code counter have been designed using existing reversible logic gate. The proposed gray code counter is designed using four no. of SAM gate, three no. of DFG, one FG and seven reversible NOT gate. The QC, GO and CI of proposed circuit are correspondingly 30, 4 and 7.

Background: In this article, we have proposed a new reversible quantum circuit block along with the quantum cost (QC), constant input (CI), garbage output (GO) and delay optimized code converterusing quantum circuit block. Method: Initially, new quantum circuit block has been designed and later reversible code converter circuits have been implemented using it. The proposed new quantum blockused to design 2’s complement code converter (2SCCC), cost efficient BCD to Excess-3 code converter (BECC) and can also be used to implement different logic functions. The QC of proposed quantum circuit block is 8. The QC and delay of the proposed 2SCCC is 8 and 1 respectively. Similarly, the QC and delay of the proposed BECC is 11 and 2 respectively. The proposed cost efficient BECC is designed using two NOT gate, one Feynman gate and one new quantum circuit block with QC is 11. Results: The improvement of QC for 2SCCC and BECC are 27.27 % and 21.43% respectively. The improvement of delay for 2SCCC and BECC are 66.67% and 50% respectively compared with respect to the latest reported results. Conclusion: So the improvement of QC and delay are very high using QCB.

The largest empty cuboid (LEC) which is axis-parallel, placed inside a 3D digital object against the background grid, captures the characteristics of the object and depicts the centrality of its shape. An efficient algorithm to determine the LEC inside a 3D digital object has been proposed here. The LEC, analogous to the largest rectangle (LR) in a 2D polygon, reveals the structure of an object in 3D. Primarily, we find the intersection polygon of the slice polygons, which are obtained from the isothetic inner cover of the digital object, at two or more consecutive levels. Starting with the bottom-most slice polygon, first we find its intersection polygon with the slice polygon at the next higher level, then the intersection of the resulting intersection polygon with the next higher level is obtained, and so on, until we reach the topmost slice. Each time, the largest rectangle inscribed inside the intersection polygon is determined and the volume of the corresponding candidate LEC is obtained by multiplying the area with the corresponding height. The above procedure is repeated starting with the slice polygon at each level. The maximum of these candidate LECs is reported. The best case computational complexity of the proposed algorithm is O(n3/2), where n is the number of voxels on the surface of the object. The experimental results on a variety of objects demonstrate that the central portion of the shape is captured by the LEC. Apart from the digital object, the algorithm can be used to find the LEC where we deal with isothetic polytope like 3D printing and modular construction.

A linear time combinatorial algorithm to construct the relative orthogonal convex hull of the inner isothetic cover of a digital object (hole-free) with respect to its outer isothetic cover is presented in this paper. The algorithm first constructs the inner and the outer isothetic covers of the digital object which is the input to the algorithm and then constructs the relative orthogonal convex hull from the inner isothetic cover and the outer isothetic cover. The algorithm can also be used to construct the relative orthogonal convex hull of one orthogonal polygon with respect to another orthogonal polygon which encloses the first polygon entirely where the input will be two orthogonal polygons one containing the other instead of the digital object. The proposed algorithm first finds the concavities of the inner polygon and checks for intruded concavities of the outer polygon. A combinatorial technique is used to obtain the relative orthogonal convex hull. The efficiency and the correctness of the algorithm are verified as it is tested on various types of digital objects.

Vessel extraction from the retinal fundus images plays a significant role in ophthalmologic disease diagnosis. Proliferative Diabetic Retinopathy (PDR) is the ultimate stage of Diabetic Retinopathy where proliferation of new and fragile blood vessels grow in human retina. These new blood vessels often show a tendency to rupture which further leads to severe damage of human eye. Neovascularization at the disk (NVD) and elsewhere (NVE) are the two general categories of PDR. So, disease diagnosis at the early stage by detecting the newly generated thin vessels demands utmost importance. Literature witness that most of the existing works emphasised on detecting only NVD. The goal of this work is to detect NVD along with NVE, as both the stages are equally devastating. The disease detection requires the extraction of vessels for subsequent analysis. A novel vessel extraction methodology has been proposed here which is capable of extracting the thick and thin vessels for further analysis. The experimental results have been tested and verified with two publicly available datasets of retinal fundus images, DRIVE and STARE. Finally, experiment for NVD and NVE detection has been carried out with DIARET-DB1 data-set. Comparison of performance with some other state-of-the-works shows superiority of the proposed methodology.

Image enhancement remains an intricate problem, crucial for image analysis. Several algorithms exist for the same. A few among these algorithms categorize images into different classes based on their statistical parameters and apply separate enhancement functions for each class. One such algorithm is the well-known adaptive gamma correction (AGC) algorithm. It works well for each class of images, but fails when the statistical parameters lie on the boundary of separation of two classes. We have developed an enhancement algorithm which can enhance images which lie on the boundary of separation equally well, as images which lie deep inside the boundary. The basic idea behind the algorithm is to combine the different enhancement functions of AGC using non-linear weight adjustments. Both contrast and brightness have been modified using these weight adjustments. We have conducted experiments on a data-set consisting of 9979 images. Results show that by using the proposed algorithm, average entropy of the enhanced images increases by 3.97% and average root mean square (rms) increases by 14.29% over AGC. Visual improvement is also perceivable.

Assessing an individual’s attentional capabilities in an interactive way with the help of EEG-based signal and utilizing it to enhance cognitive control in obsessive–compulsive disorder is the primary focus of this paper. To realize this objective, BCI technology is used by studying the EEG signals emitted by the brain and processing the data in real time. In this paper, we presented the development of a BCI-based gaming application with a mode of interactive media that takes advantage of the EEG signal gathered from the hardware used. Further, we discussed the scope of enhancing cognitive control capabilities in various psychiatric disorders in an interactive way with the help of the developed application.

Background: Over the last few decades, reversible logic system/circuits have received considerable attention in the diversified fields such as nanotechnology, quantum computing, cryptography, optical computing and low power design of VLSI circuits due to their low power dissipation characteristics. Methods: In this paper, we proposed the design of reversible shift register (SR) i.e. serial-in-serial out (SISO), serial-in-parallel out (SIPO), parallel-in-serial out (PISO) and parallel-in-parallel out (PIPO) SR using a reduced number of reversible logic gates and garbage output. Results: As compared to previously reported results, the improvement in our proposed model of SISO, SIPO, PISO and PIPO was found to be 50 – 66.66 %, 42.85 – 66.66 %, 12.5 – 53.33 % and 50 – 66.66 % respectively, in terms of the number of reversible logic gates.

The internet of things (IoT) is now an in-demand technology that has been adopted in various applications and includes various embedded devices, sensors and other objects connected to the Internet. Due to the rapid development of this technology, it covers a significant portion of the research interests nowadays. IoT devices are typically designed for collecting different types of data from various sources and transmitting them in digitized form. However, data security is the burning issue in the IoT technology, which can broadly impact the privacy of crucial data. In this regard, a new lightweight encryption method called LRBC has been proposed in this work for resource constraint IoT devices which can provide data security at the sensing level. The LRBC has used the structural advantages of both substitution–permutation network (SPN) and Feistel structure together to achieve better security. Furthermore, the proposed method has been tested on NEXYS 4 DDR FPGA (Artix-7) trainer kit and implemented for application specific integrated circuit (ASIC) chip on TSMC 65 nm technology. The proposed algorithm consumes very less power of 11.40 μW and occupies a 258.9 GE (Gate Equivalent) area. Besides, a thorough security analysis shows that the proposed scheme ensures high security against various attacks with robustness. Moreover, the average avalanche effect of LRBC is found to be 58% and 55.75% concerning plaintext and key, respectively.

For biometric identification with retina, vascular structure based features bear significance in preparing retinal digital templates. This requires analysis of large amount of data from different sources. It needs a faster and robust automated system to extract the quantitative measures from huge amount of retinal images. Therefore, fast and accurate detection of existing retinal features is an important element for a successful biometric identification. In this work, we propose to design and implement a retinal biometric key generation framework with deep neural network. The purpose is to replace the semi-automated or automated retinal vascular feature identification methods. The approach begins with segmentation from coloured fundus images, followed by selection of some unique features like center of optic disc, macula center and distinct bifurcation points on a convolutional neural network model. For better understanding, the key generation process has finally been shown with the help of a graphical user interface. This network was trained and tested with the training and testing images of DRIVE dataset and some of our previously published result sets on automated feature extraction methods. The network was trained on NVIDIA Titan Xp GPU provided by NVIDIA corporation.

A short review on avalanche transit time devices has been presented in this chapter. From its first proposal till the state of the art, impact avalanche transit time (IMPATT) sources for the generation of terahertz (THz) waves have been briefly described. The noise outputs of the sources at different electromagnetic spectrum have been discussed. The details of deferent device structures, potential materials, popular simulation techniques, etc. are also appended in this chapter.

Terahertz radiation has drawn enormous attention in recent times due to its various application possibilities. This chapter reviews various emerging trends and well-established technologies in Terahertz biomedical. Due to its extraordinary sensing capabilities, non-invasive, non-ionizing properties, sensitive instrumentations for spectroscopy and imaging, Terahertz has found various biomedical sensing applications from biomolecules, proteins to cells and tissues. This chapter highlights terahertz device engineering, system technologies, range of materials, aiming at various biomedical applications. It also includes emerging topics such as terahertz biomedical imaging, pattern recognition and tomographic reconstruction by machine learning and artificial intelligence, for possible biomedical imaging applications.

Great demand of suitable terahertz (THz) sources requires optimized design and appropriate simulation tool for designing and investigating the device capability at the design spectrum before actual fabrication. Design and simulation of complex devices like THz avalanche transit time (ATT) diodes require comprehensive and accurate simulation model for small-signal, large-signal and noise simulation before the actual fabrication. This chapter is devoted to summarize these three types of simulation methodologies for THz ATT sources.

In this preparatory chapter, a brief introduction to the generation, detection and processing of terahertz (THz) signals is given. Short descriptions of prospective applications of THz signals as well as some state-of-the-art electronic and photonic devices have also been included in this chapter. A chapter-wise overview of the entire book has been incorporated at the end of this introductory chapter.

As the amount of user-generated textual content grows rapidly, text summarization algorithms are increasingly being used to provide users a quick overview of the information content. Traditionally, summarization algorithms have been evaluated only based on how well they match human-written summaries (e.g. as measured by ROUGE scores). In this work, we propose to evaluate summarization algorithms from a completely new perspective that is important when the user-generated data to be summarized comes from different socially salient user groups, e.g. men or women, Caucasians or African-Americans, or different political groups (Republicans or Democrats). In such cases, we check whether the generated summaries fairly represent these different social groups. Specifically, considering that an extractive summarization algorithm selects a subset of the textual units (e.g. microblogs) in the original data for inclusion in the summary, we investigate whether this selection is fair or not. Our experiments over real-world microblog datasets show that existing summarization algorithms often represent the socially salient user-groups very differently compared to their distributions in the original data. More importantly, some groups are frequently under-represented in the generated summaries, and hence get far less exposure than what they would have obtained in the original data. To reduce such adverse impacts, we propose novel fairness-preserving summarization algorithms which produce high-quality summaries while ensuring fairness among various groups. To our knowledge, this is the first attempt to produce fair text summarization, and is likely to open up an interesting research direction., Comment: This work has been accepted for presentation at the ACM Conference on Computer-Supported Cooperative Work and Social Computing (CSCW) 2019

There are many approaches exist in literature to register retinal images from their RGB fundus stages. Previous approaches need to take special care upon transformations like rotation and translation. In this paper, an image registration method has been proposed which is robust against transformation and does not need extra burden to handle it. In particular, our proposed approach consists of three major stages. At the first stage, vascular network has been segregated into many other coloured sub-categories on the segmented fundus image. The second stage is to find out arterial and venous segments among the coloured sub-categories of the network by following the proposed artery-vein (A/V) classification method. Finally, each arterial and venous segments are combined together to form arterial and venous subtrees which resembles a binary tree. Finding inorder traversal of each such subtrees complete the process for image registration. We evaluate the approach with ground truth images of a public database called RITE. The work shows an arterial-venous classification rate as $$89\%$$ using an automated vessel segmented image as input which is comparable with other existing methods.

Spectral unmixing is one of the post processing operation of hyperspectral image processing. In general, it is observed that the traditional algorithms are not efficient enough to estimate the abundance fractions of endmembers. In recent years, neural network based approaches are competent enough to perform the complex operations for remote sensing applications. After the origination of deep learning, number of deep learning approaches have been proposed and trained using different loss functions for performing several complex operations. Hence, it is necessary to select suitable deep learning model and loss function for the application(s). The objective of this work is to study the suitability of the model which affects the performance of the unmixing operation. In this work, estimation accuracy of different deep learning models and loss functions for spectral unmixing operation using hyperspectral data have been studied. In this work, five deep learning models are implemented and trained using four loss functions. Evaluation of proposed study has been carried using available real hyperspectral datasets. Hence, It is observed from the study that the different deep learning models and loss functions affect the estimation accuracy of spectral abundance fractions. In this work, parallel convolution 1-D model has been performed best among the implemented approaches for estimation of abundance fractions.

To fight against the present pandemic scenario of COVID-19 outbreak, medication with drugs and vaccines is extremely essential other than ventilation support. In this paper, we present a list of ligands which are expected to have the highest binding affinity with the S-glycoprotein of 2019-nCoV and thus can be used to make the drug for the novel coronavirus. Here, we implemented an architecture using 1D convolutional networks to predict drug–target interaction (DTI) values. The network was trained on the KIBA (Kinase Inhibitor Bioactivity) dataset. With this network, we predicted the KIBA scores (which gives a measure of binding affinity) of a list of ligands against the S-glycoprotein of 2019-nCoV. Based on these KIBA scores, we are proposing a list of ligands (33 top ligands based on best interactions) which have a high binding affinity with the S-glycoprotein of 2019-nCoV and thus can be used for the formation of drugs.

Early and accurate detection of malaria is the main key in controlling and eradicating this deadly disease. An efficient automated tool for examining stained blood slides can be very helpful in this regard. This paper proposes an approach for automatic detection of malarial parasites from thin blood smears. Texture- and intensity-based analyses are performed to detect the blood particles in the pre-processed image. Extracted red blood cell particles are then sent to a convolutional neural network for further probing. The proposed CNN architecture is trained with a publicly available dataset along with some Giemsa-stained blood smear images collected from a hospital. The performance of malaria detection process gives a satisfactory dice score of 0.95.

This work proposes a method of polygonization of the surface of triangulated 3D digital objects by merging the edge-adjacent face triangles having similar norms. Least square method is applied to the boundary vertices of the set of the merged face triangles, which yields a new norm optimizing the norms of the individual face triangles constituting the merged set. The coordinates of the boundary vertices are recalculated with respect to this new norm. Finally, all the instances of each vertex is averaged to obtain the final coordinates. In this process, many of the vertices of the original input object are dropped, giving rise to a number of polygons describing the object, which is substantially less than the number of original face triangles. Thus, a considerable amount of compression is achieved, Also, polygons representing the object can reveal its structure and shape and are amenable to improved shape analysis. Contribution: It proposes an algorithm to polygonize 3D digital object surface achieving significant compression. Contribution: It proposes an algorithm to polygonize 3D digital object surface achieving significant compression.

The following research paper gives a different type of voltage mode multifunction filter which is of first order, which consist of only one OTRA block that is operational transresistance amplifier and few of the passive components. This structure can be used for employing different types of filter functions of 1st order. The proposed device is simulated with the parameters of 0.5µm CMOS which is defined by MOSIS (AGILENT). The results are verified using simulation software PSPICE.

Predicting future trajectories of agents in a dynamic environment is essential for natural and safe decision making of autonomous agents. The trajectory of an agent in such an environment not only depends on the past motion of that agent but also depends on its interaction with other agents present in that environment. To capture the effect of other agents on trajectory prediction of an agent, we propose a three stream topology-aware graph convolutional network (TAGCN) for interaction message passing between the agents. In addition, temporal encoding of local- and global-level topological features are fused to better characterize dynamic interactions between participants over time. Results are competitive compared to previous best methods for trajectory prediction on ETH and UCY datasets and highlights the need for both local and global interaction structure.

This paper proposes an efficient method to compute entropy and joint entropy of images. Entropy of images is used to determine its quality. It is defined as the randomness or uncertainty present in the image. Similarly, joint entropy is a measure of the uncertainty present in the overlapped region of two images. Entropy and joint entropy computations are vital in several image processing applications. Intensity based image registration is done by maximizing the mutual information between two images. Mutual information is nothing but the difference between sum of individual entropies and joint entropy of two images. Image registration has applications, especially in the medical field, e.g. diagnosis and treatment of diseases. The entropy and joint entropy computation methods proposed in this paper are computationally less expensive than the standard methods. Entropy computation takes 78.60% less time than the standard method while computational time of joint entropy is reduced by 83.59%. This increase in efficiency comes at the cost of an error of 1.52% in entropy and 4.54% in joint entropy.

Bone cancer originates from bone and rapidly spreads to the rest of the body affecting the patient. A quick and preliminary diagnosis of bone cancer begins with the analysis of bone X-ray or MRI image. Compared to MRI, an X-ray image provides a low-cost diagnostic tool for diagnosis and visualization of bone cancer. In this paper, a novel technique for the assessment of cancer stage and grade in long bones based on X-ray image analysis has been proposed. Cancer-affected bone images usually appear with a variation in bone texture in the affected region. A fusion of different methodologies is used for the purpose of our analysis. In the proposed approach, we extract certain features from bone X-ray images and use support vector machine (SVM) to discriminate healthy and cancerous bones. A technique based on digital geometry is deployed for localizing cancer-affected regions. Characterization of the present stage and grade of the disease and identification of the underlying bone-destruction pattern are performed using a decision tree classifier. Furthermore, the method leads to the development of a computer-aided diagnostic tool that can readily be used by paramedics and doctors. Experimental results on a number of test cases reveal satisfactory diagnostic inferences when compared with ground truth known from clinical findings.

In this paper we propose a set of novel regression models for readability scoring in Bengali language, which can also be used for Hindi, making use of several lexical, surface-level, syntac...

The extraction of structural information from the vascular network captured in a retinal image is an essential component of computer-aided diagnosis for many ophthalmological, cardiovascular, and systemic disorders. Although several methods have been reported in the past to tackle this problem, ensuring clarity and high accuracy while segmenting various objects from a retinal image still poses a challenge. In this work, we have proposed a novel technique to extract the underlying structure of blood vessels from a retinal image that includes segmentation of optic disc region and identification of veins, arteries, and bifurcation points in the vascular network. Unlike previous approaches, the proposed method relies mostly on the processing of digital-geometric features and integrates them with conventional image analysis for effective segmentation of various objects in the vascular network. The method outperforms several prior work in terms of segmentation accuracy, and experiments on several retina...

Authorship Attribution is a long-standing problem in Natural Language Processing. Several statistical and computational methods have been used to find a solution to this problem. In this article, we have proposed methods to deal with the authorship attribution problem in Bengali. More specifically, we proposed a supervised framework consisting of lexical and shallow features and investigated the possibility of using topic-modeling-inspired features, to classify documents according to their authors. We have created a corpus from nearly all the literary works of three eminent Bengali authors, consisting of 3,000 disjoint samples. Our models showed better performance than the state-of-the-art, with more than 98% test accuracy for the shallow features and 100% test accuracy for the topic-based features. Further experiments with GloVe vectors [Pennington et al. 2014] showed comparable results, but flexible patterns based on content words and high-frequency words [Schwartz et al. 2013] failed to perform as well as expected.

This work presents a novel approach to classify digital objects based on outer isothetic cover (OIC). The descriptor vectors of the isothetic cover are used as the shape descriptors of a particular digital object. Two objects may be of similar kind of shape if their shape descriptors are analogous. To find a match between two descriptors, common substrings are found out and based on their length of the common substring a decision is made. Experimental results show encouraging results.

Thick blood smear examination essentially plays an important role in rapid screening of malaria parasite. In this paper, an automatic malaria parasite detector is proposed to perceive the malaria-infected blood cells in a stained thick blood smear image. The detector hence can verify presence of malarial parasites in blood cells and count the number against each white blood corpuscle. It could precisely accelerate the process of diagnosing malaria in human blood and thus reduces the chances of human errors. Here, an automated image segmentation technique has been developed that proposes to efficiently improve the accuracy over manual techniques. The experimental results show that the proposed method can provide impressive performance in segmenting the malarial parasites from a thick blood smear image on some published malaria dataset and also taken under an origami-based paper microscope. The algorithm makes use of certain preprocessing methods for comparatively better results in the proposed intensity-based segmentation technique, followed by a dual threshold band-based technique. The experimental results are comparable to other existing and traditional malaria segmentation processes.