Sabyasachi Patra
@iiit-bh.ac.in
Assistant Professor, Department of CSE
International Institute of Information Technology Bhubaneswar
Scopus Publications
Scopus Publications
Subham Kumar Padhy, Anjali Mohapatra, and Sabyasachi Patra
Springer Nature Singapore
Tushar Ranjan Sahoo, Sabyasachi Patra, and Swati Vipsita
Elsevier BV
Tushar Ranjan Sahoo, Swati Vipsita, and Sabyasachi Patra
Springer Science and Business Media LLC
Subham Kumar Padhy, Saurav Mahalik, Himanshu Nayak, Sabyasachi Patra, and Anjali Mohapatra
IEEE
Abnormal heart rhythm or irregular heartbeat, often known as arrhythmia. It is a kind of cardiovascular illness that necessitates a precise and fast diagnosis. Because of its simplicity and non-invasive nature, an electrocardiogram (ECG) that detects the electric activity of the heart has been frequently used to identify cardiac disorders. Each heartbeat's electrical signal, the peak of action impulse waveforms produced by various specialised cardiac tissues, can be used to diagnose various heart defects. Deep learning has evolved as a significant technique in recent decades due to its ability to handle vast amounts of data. The use of hidden layers in the convolution layer have enhanced pattern recognition performance. Deep learning has aided in the automation of medical image analysis and can aid in detecting of any anomalies in the medical images. In this work, ECG-based automated irregular heartbeat prediction is conducted to determine to which arrhythmia class it belongs with greater accuracy and less data loss. This study is based on convolutional neural networks, which are used to evaluate ECG images. For the normal case and cases impacted by various arrhythmias and my-ocardial infarction, the signals correspond to electrocardiogram (ECG) forms of heartbeats.lD-CNN, ResNet34, ResNet50, vgg16, and vgg19 models are utilised to predict of cardiac arrhythmia. vgg16 performed the best and is chosen to be further tweaked to improve accuracy to 99.79 percent.
Subham Kumar Padhy, Anjali Mohapatra, and Sabyasachi Patra
IEEE
The classification of heart sounds is of utmost importance in promptly identifying cardiovascular disorders, particularly in small primary healthcare clinics. Although significant advancements have been achieved in the classification of heart sounds recently, most of these developments rely on traditional segmented attributes and classifiers with limited depth. These traditional approaches to representing and classifying acoustic signals must be revised to capture the nuances of heart adequately sounds. They often face challenges in delivering accurate results due to the cardiac environment's complex and variable acoustic conditions. This study suggests an enhanced Mel-Frequency Cepstrum Coefficient (MFCC) feature-based technique for classifying heart sounds and a Long-Short Term Memory neural network (LSTM). The neural network receives MFCC-based features to perform feature learning, followed by the classification task. The experiment's findings show that the suggested technique performs effectively over a range of tolerance windows.
Soubhagya Ranjan Mallick, Veena Goswami, Rakesh Kumar Lenka, Sabyasachi Patra, Vinay Kumar, and Rabindra K. Barik
IEEE
Regarding geographic cloud computing, the three most crucial services are storage, processing, and visualisation. This cloud computing setup necessitates using the Internet of Spatial Things (IoST) technology. Users, planners, and environmentalists can all benefit from this technology because it can be used to build geospatially enabled smart environmental monitoring and assessment frameworks. There are many technical challenges in the IoST-enabled geospatial cloud environment, including scalability, security, privacy, and interoperability. Timely provision of real-time services is crucial for most environmental monitoring systems. The present work proposes a novel priority Queueing assisted IoST-Fog-Blockchain framework for the environmental monitoring and assessment system to resolve these issues. Temporarily storing geographical data, processing geospatial data, and visualising geospatial data are all services provided by the proposed framework, all of which can be accessed from the closest fog node. It minimises the Geospatial cloud servers' overload, traffic density, bandwidth requirements and service time. A novel priority Queueing analytical approach implies enhancing access control and ensuring the timely delivery of these services. Finally, it illustrates the several numerical outcomes that show how the proposed priority analytical model may be used to anticipate the performance indices of the presented system.
Tushar Ranjan Sahoo, Swati Vipsita, and Sabyasachi Patra
Inderscience Publishers
Sabyasachi Patra
Elsevier BV
Bhawani Sankar Biswal, Sabyasachi Patra, Anjali Mohapatra, and Swati Vipsita
Institution of Engineering and Technology (IET)
Computational analysis of microarray data is crucial for understanding the gene behaviours and deriving meaningful results. Clustering and biclustering of gene expression microarray data in the unsupervised domain are extremely important as their outcomes directly dominate healthcare research in many aspects. However, these approaches fail when the time factor is added as the third dimension to the microarray datasets. This three-dimensional data set can be analysed using triclustering that discovers similar gene sets that pursue identical behaviour under a subset of conditions at a specific time point. A novel triclustering algorithm (TriRNSC) is proposed in this manuscript to discover meaningful triclusters in gene expression profiles. TriRNSC is based on restricted neighbourhood search clustering (RNSC), a popular graph-based clustering approach considering the genes, the experimental conditions and the time points at an instance. The performance of the proposed algorithm is evaluated in terms of volume and some performance measures. Gene Ontology and KEGG pathway analysis are used to validate the TriRNSC results biologically. The efficiency of TriRNSC indicates its capability and reliability and also demonstrates its usability over other state-of-art schemes. The proposed framework initiates the application of the RNSC algorithm in the triclustering of gene expression profiles.
Sabyasachi Patra and Anjali Mohapatra
Elsevier BV
Sabyasachi Patra and Anjali Mohapatra
Institution of Engineering and Technology (IET)
Network motifs are recurrent and over‐represented patterns having biological relevance. This is one of the important local properties of biological networks. Network motif discovery finds important applications in many areas such as functional analysis of biological components, the validity of network composition, classification of networks, disease discovery, identification of unique subunits etc. The discovery of network motifs is a computationally challenging task due to the large size of real networks, and the exponential increase of search space with respect to network size and motif size. This problem also includes the subgraph isomorphism check, which is Nondeterministic Polynomial (NP)‐complete. Several tools and algorithms have been designed in the last few years to address this problem with encouraging results. These tools and algorithms can be classified into various categories based on exact census, mapping, pattern growth, and so on. In this study, critical aspects of network motif discovery, design principles of background algorithms, and their functionality have been reviewed with their strengths and limitations. The performances of state‐of‐art algorithms are discussed in terms of runtime efficiency, scalability, and space requirement. The future scope, research direction, and challenges of the existing algorithms are presented at the end of the study.
Srikant Kumar, Anjali Mohapatra, Sabyasachi Patra, and Sunakshi Mamgain
IEEE
Questions which are syntactically different, yet having the same intent, give a poor encounter to both the writer of an answer as well as the individual who searches for the answer in social Q&A online platforms such as Quora, Yahoo Answers, and StackOverflow. In order to maintain a rich and diverse database of answers, ensuring the uniqueness of every question on such Q&A platforms is of utmost necessity. The objective of this proposed work is to eliminate redundancy/duplicacy of userentered questions so as to increase the relevance of the answer(s) provided to semantically similar questions. This problem is a closed challenge taken from Kaggle, an online platform to learn and compete in data science challenges. The dataset worked upon in this paper has been obtained from Kaggle and LSTM with Euclidean Distance outperform other algorithms with log loss of 0.14.
Sabyasachi Patra and Anjali Mohapatra
Institution of Engineering and Technology (IET)
The biological network plays a key role in protein function annotation, protein superfamily classification, disease diagnosis, etc. These networks exhibit global properties like small‐world property, power‐law degree distribution, hierarchical modularity, robustness, etc. Along with these, the biological network also possesses some local properties like clustering and network motif. Network motifs are recurrent and statistically over‐represented subgraphs in a target network. Operation of a biological network is controlled by these motifs, and they are responsible for many biological applications. Discovery of network motifs is a computationally hard problem and involves a subgraph isomorphism check which is NP‐complete. In recent years, researchers have developed various tools and algorithms to detect network motifs efficiently. However, it is still a challenging task to discover the network motif within a practical time bound for the large motif. In this study, an efficient pattern‐join based algorithm is proposed to discover network motif in biological networks. The performance of the proposed algorithm is evaluated on the transcription regulatory network of Escherichia coli and the protein interaction network of Saccharomyces cerevisiae. The running time of the proposed algorithm outperforms most of the existing algorithms to discover large motifs.
Sabyasachi Patra and Anjali Mohapatra
PeerJ
Network motifs play an important role in the structural analysis of biological networks. Identification of such network motifs leads to many important applications such as understanding the modularity and the large-scale structure of biological networks, classification of networks into super-families, and protein function annotation. However, identification of large network motifs is a challenging task as it involves the graph isomorphism problem. Although this problem has been studied extensively in the literature using different computational approaches, still there is a lot of scope for improvement. Motivated by the challenges involved in this field, an efficient and scalable network motif finding algorithm using a dynamic expansion tree is proposed. The novelty of the proposed algorithm is that it avoids computationally expensive graph isomorphism tests and overcomes the space limitation of the static expansion tree (SET) which makes it enable to find large motifs. In this algorithm, the embeddings corresponding to a child node of the expansion tree are obtained from the embeddings of a parent node, either by adding a vertex or by adding an edge. This process does not involve any graph isomorphism check. The time complexity of vertex addition and edge addition are O(n) and O(1), respectively. The growth of a dynamic expansion tree (DET) depends on the availability of patterns in the target network. Pruning of branches in the DET significantly reduces the space requirement of the SET. The proposed algorithm has been tested on a protein–protein interaction network obtained from the MINT database. The proposed algorithm is able to identify large network motifs faster than most of the existing motif finding algorithms.
Sabyasachi Patra and Anjali Mohapatra
World Scientific Pub Co Pte Lt
Networks are powerful representation of topological features in biological systems like protein interaction and gene regulation. In order to understand the design principles of such complex networks, the concept of network motifs emerged. Network motifs are recurrent patterns with statistical significance that can be seen as basic building blocks of complex networks. Identification of network motifs leads to many important applications, such as understanding the modularity and the large-scale structure of biological networks, classification of networks into super-families, protein function annotation, etc. However, identification of network motifs is challenging as it involves graph isomorphism which is computationally hard. Though this problem has been studied extensively in the literature using different computational approaches, we are far from satisfactory results. Motivated by the challenges involved in this field, an efficient and scalable network Motif Discovery algorithm based on Expansion Tree (MODET) is proposed. Pattern growth approach is used in this proposed motif-centric algorithm. Each node of the expansion tree represents a non-isomorphic pattern. The embeddings corresponding to a child node of the expansion tree are obtained from the embeddings of the parent node through vertex addition and edge addition. Further, the proposed algorithm does not involve any graph isomorphism check and the time complexities of these processes are [Formula: see text] and [Formula: see text], respectively. The proposed algorithm has been tested on Protein–Protein Interaction (PPI) network obtained from the MINT database. The computational efficiency of the proposed algorithm outperforms most of the existing network motif discovery algorithms.
Sabyasachi Patra and Anjali Mohapatra
IEEE
Biological networks such as gene regulatory network, metabolic network and protein interaction network are extensively studied in the literature since last two decades. The various concept of graph theory is widely used to extract biological information from these networks, such as prediction of biological function, detection of protein complexes, the discovery of new interactions, diagnosis of disease, and drug design etc. Network motif analysis is one of the important approaches for functional analysis in the biological network. However, the contribution of biological elements towards these motifs is not clearly defined. Most of the literature discussed the biological significance of motifs as a whole. In this manuscript, the role of proteins for each identified motif is defined in an interaction network. These roles are concatenated to form a motif feature vector. The agglomerative hierarchical clustering algorithm is applied for clustering of proteins based on the above-identified feature vectors. Clustering of proteins leads to many application like protein superfamily classification, protein function annotation etc. The proposed method is evaluated on the protein interaction data of Human herpesvirus-1, Human herpesvirus-8 and Escherichia coli from the MINT database. The performance of the proposed clustering algorithm is evaluated by using the cophenetic correlation coefficient. Cophenetic correlation coefficients of all the output clusters are almost close to 1 which indicates their high quality.
Sabyasachi Patra and Subhendu Kumar Acharya
IEEE
Speaker identification based on speech signal has been receiving enhanced attention from the research community. In this context the effect of dimension reduction of feature vectors using Principal Component Analysis (PCA) and Weighted Principal Component Analysis (WPCA) are compared for speaker identification in a noisy environment. MFCC feature vectors are used as original features and their dimension is reduced by PCA and WPCA techniques and then evaluated by GMM classifier. Speaker identification rate is calculated under different SNR to test the robustness of the speaker identification system. In low SNR, the speaker identification rate becomes double after reducing the dimension of feature vectors by 50% as compared to original one. The performance of WPCA is 10% to 20% better than PCA under different SNR.