Open Bioinformatics Journal最新文献

{"title":"Decision-making Support System for Predicting and Eliminating Malnutrition and Anemia","authors":"Manasvi Jagadeesh Maasthi, Harinahalli Lokesh Gururaj, Vinayakumar Ravi, Basavesha D, Meshari Almeshari, Yasser Alzamil","doi":"10.2174/0118750362246898230921054021","DOIUrl":"https://doi.org/10.2174/0118750362246898230921054021","url":null,"abstract":"Aims: This study investigates predicting and eliminating malnutrition and anemia using ML Algorithms and comparing all the methods with various malnutrition-based parameters. Background: The health of the nation is more important than the wealth of the nation. Malnutrition and anemia are serious health issues but the least importance is given to eradicate them. Objective: Proper nutrition is an essential component for the survival, growth, and development of infants, children, and women who in turn give birth to infants. Methods: In the proposed system, machine learning approaches are utilized to predict the malnutrition status of children under five years of age and anemia in men and women using old datasets and further providing a suitable diet recommendation to overcome the disease. Classification techniques are being used for malnutrition as well as anemia prediction. Results: Algorithms such as Naïve Bayes classifier (NBC), Decision Tree (DT) algorithm, Random Forest (RF) algorithm, and K-Nearest Neighbor (k-NN) algorithm are utilized for prediction. The results obtained by these algorithms are 94.47%, 85%, 95.49%, and 63.15%. When compared, Naïve Bayes and random forest algorithm provided efficient results for malnutrition and anemia, respectively. Conclusion: By testing and validating, preventive actions can be taken with the help of medical experts and dieticians to reduce malnutrition and anemia conditions among children and elders, respectively.","PeriodicalId":38956,"journal":{"name":"Open Bioinformatics Journal","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-10-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136318383","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Immunoinformatics Approach for the Design of Chimeric Vaccine Against Whitmore Disease","authors":"Shalini Maurya, Salman Akhtar, Mohammad Kalim Ahmad Khan","doi":"10.2174/0118750362253383230922100803","DOIUrl":"https://doi.org/10.2174/0118750362253383230922100803","url":null,"abstract":"Purpose: Multidrug-resistant Burkholderia pseudomallei is associated with significant morbidity and mortality. Hence, there is a requirement for a vaccine for this pathogen. Using subtractive proteomics and reverse vaccinology approaches, we have designed a chimeric multiepitope vaccine against the pathogen in the present study. Methods: Twenty-one non-redundant pathogen proteomes were mined using a subtractive proteomics strategy. Out of these, by various analyses, we found proteins that were non-homologous to humans, essential, and virulent. BLASTp against the PDB database and Pocket druggability analysis yielded nine proteins whose 3D structure is available and are druggable. Four proteins that could be candidates for vaccines were identified by subcellular localization and antigenicity prediction, and they could be used in reverse vaccinology methods to create a chimeric multiepitope vaccine. Results: Using online resources and servers, MHC class I, II, and B cell epitopes were identified. The predicted epitopes were selected based on analysis of toxicity, solubility, allergenicity, and hydrophilicity. These predicted epitopes, which were immunogenic, were used for the construction of a multivalent chimeric vaccine. The epitopes, adjuvants, linkers, and PADRE amino acid sequences were employed to create the vaccine. Shortlisted vaccine constructs also interact with the HLA allele and TLR4, as evident from docking and molecular dynamics simulation. Thus, vaccine construct V1 can elicit an immune response against Burkholderia pseudomallei . Conclusion: The availability of the proteome of B. pseudomallei has made this study possible through the usage of various in silico approaches. We could shortlist vaccine targets using subtractive proteomics and then construct chimeric vaccines using reverse vaccinology and immunoinformatics approaches.","PeriodicalId":38956,"journal":{"name":"Open Bioinformatics Journal","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-10-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135666306","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A New Deep Learning Model based on Neuroimaging for Predicting Alzheimer's Disease","authors":"Kiran P., Sudheesh K. V., Vinayakumar Ravi, Meshari Almeshari, Yasser Alzamil, Sunil Kumar D. S., Harshitha R.","doi":"10.2174/0118750362260635230922051326","DOIUrl":"https://doi.org/10.2174/0118750362260635230922051326","url":null,"abstract":"Background: The psychological aspects of the brain in Alzheimer's disease (AD) are significantly affected. These alterations in brain anatomy take place due to a variety of reasons, including the shrinking of grey and white matter in the brain. Magnetic resonance imaging (MRI) scans can be used to measure it, and these scans offer a chance for early identification of AD utilizing classification methods, like convolutional neural network (CNN). The majority of AD-related tests are now constrained by the test measures. It is, thus, crucial to find an affordable method for image categorization using minimal information. Because of developments in machine learning and medical imaging, the field of computerized health care has evolved rapidly. Recent developments in deep learning, in particular, herald a new era of clinical decision-making that is heavily reliant on multimedia systems. Methods: In the proposed work, we have investigated various CNN-based transfer-learning strategies for predicting AD using MRI scans of the brain's structural organization. According to an analysis of the data, the suggested model makes use of a number of sites related to Alzheimer's disease. In order to interpret structural brain pictures in both 2D and 3D, the Alzheimer's Disease Neuroimaging Initiative (ADNI) dataset includes straightforward CNN designs based on 2D and 3D convolutions. Results: According to these results, deep neural networks may be able to automatically learn which imaging biomarkers are indicative of Alzheimer's disease and exploit them for precise early disease detection. The proposed techniques have been found to achieve an accuracy of 93.24%. Conclusion: This research aimed to classify Alzheimer's disease (AD) using transfer learning. We have used strict pre-processing steps on raw MRI data from the ADNI dataset and used the AlexNet, i.e ., Alzheimer's disease has been categorized using pre-processed data and the CNN classifier.","PeriodicalId":38956,"journal":{"name":"Open Bioinformatics Journal","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-10-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136183087","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Early Prediction of Covid-19 Samples from Chest X-ray Images using Deep Learning Approach","authors":"K V Sudheesh, None Kiran, Harinahalli Lokesh Gururaj, Vinayakumar Ravi, Meshari Almeshari, Yasser Alzamil","doi":"10.2174/18750362-v16-231005-2023-5","DOIUrl":"https://doi.org/10.2174/18750362-v16-231005-2023-5","url":null,"abstract":"Aims: In this study, chest X-ray (CXR) and computed tomography (CT) images are used to analyse and detect COVID-19 using an unsupervised deep learning-based feature fusion approach. Background: The reverse transcription-polymerase chain reaction (RT-PCR) test, which has a reduced viral load, sampling error, etc., is used to detect COVID-19, which has sickened millions of people worldwide. It is possible to check chest X-rays and computed tomography scans because the majority of infected persons have lung infections. The COVID-19 diagnosis can be made early using both CT and CXR imaging modalities, which is an alternative to the RT-PCR test. Objective: The manual diagnosis of CXR pictures and CT scans is labor and time-intensive. Many AI-based solutions are being investigated to tackle this problem, including deep learning-based detection models, which can be utilized to assist the radiologist in making a more accurate diagnosis. However, because of the demand for specialized knowledge and high annotation costs, the amount of annotated data available for COVID-19 identification is constrained. Additionally, the majority of current cutting-edge deep learning-based detection models use supervised learning techniques. Because a tagged dataset is not required, we have investigated various unsupervised learning models for COVID-19 identification in this work. Methods: In this study, we suggest a COVID-19 detection method based on unsupervised deep learning that makes use of the feature fusion technique to improve performance. Based on this an automated CNN model is built for the detection of COVID-19 samples from healthy and pneumonic cases using chest X-ray images. Results: This model has scored an accuracy of about 99% for the classification between covid positive and covid negative. Based on this result further classification will be done for pneumonic and non-pneumonic which has scored an accuracy of 94%. Conclusion: On both datasets, the COVID-19 detection method based on feature fusion and deep unsupervised learning showed promising results. Additionally, it outperforms four well-known unsupervised methods already in use.","PeriodicalId":38956,"journal":{"name":"Open Bioinformatics Journal","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-10-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135922821","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Electronic Health Record (EHR) System Development for Study on EHR Data-based Early Prediction of Diabetes Using Machine Learning Algorithms","authors":"Jagadamba G, Shashidhar R, Gururaj H L, Vinayakumar Ravi, Meshari Almeshari, Yasser Alzamil","doi":"10.2174/18750362-v16-e230906-2023-15","DOIUrl":"https://doi.org/10.2174/18750362-v16-e230906-2023-15","url":null,"abstract":"Aims: This research work aims to develop an interoperable electronic health record (EHR) system to aid the early detection of diabetes by the use of Machine Learning (ML) algorithms. A decision support system developed using many ML algorithms results in optimizing the decision in preventive care in the health information system. Methods: The proposed system consisted of two models. The first model included interoperable EHR system development using a precise database structure. The second module comprised of data extraction from the EHR system, data cleaning, and data processing and prediction. For testing and training, about 1080 patients’ health record was considered. Among 1080, 1000 records were from the Kaggle dataset, and 80 records were demographic information from patients who visited our health center of Siddaganga organization for a regular checkup or during emergencies. The demographic information was collected from the proposed EHR system. Results: The proposed system was tested for the interoperability nature of the EHR system and accuracy in diabetic disease prediction using the proposed decision support system. The proposed EHR system development was tested for interoperability by random updations from various systems maintained in the laboratory. Each system acted like the admin system of different hospitals. The EHR system was tested for handling the load and interoperability by considering user view status, system matching with the real world, consistency in data updations, security etc . However, in the prediction phase, diabetes prediction was concentrated. The features considered were not randomly chosen; however, the features were those prescribed by a doctor who insisted that the features were sufficient for initial prediction. The reports collected from the doctors revealed several features they considered before giving the test details. The proposed system dataset was split into test and train datasets with eight proper features taken as input and one set as a target variable where the result was present. After this, the model was imported using standard “sklearn” libraries, and it fit with the required number of estimators, that is, the number of decision trees. The features included pregnancies, glucose level, blood pressure, skin thickness, insulin level, bone marrow index, diabetic pedigree function, age, weight, etc . At the outset, the research work concentrated on developing an interoperable EHR system, identifying the expectation of diabetic and non-diabetic conditions and demonstrating the accuracy of the system. Conclusion: In this study, the first aim was to design an interoperable EHR system that could help in accumulating, storing, and sharing patients' timely health records over a lifetime. The second aim was to use EHR data for early prediction of diabetes in the user. To confirm the accuracy of the system, the system was tested regarding interoperability to support early prediction through a decision supp","PeriodicalId":38956,"journal":{"name":"Open Bioinformatics Journal","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-10-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135547188","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Comparative Functional Genomics Studies for Understanding the Hypothetical Proteins in\u0000 Mycobacterium Tuberculosis Variant Microti 12","authors":"Tejaswini Vijay Shinde, Tejas Gajanan Shinde, V. V. Chougule, Anagha Rajendra Ghorpade, Geeta Vikas Utekar, Amol S Jadhav, Bandu Shamlal Pawar, S. Sanmukh","doi":"10.2174/18750362-v16-e230711-2023-2","DOIUrl":"https://doi.org/10.2174/18750362-v16-e230711-2023-2","url":null,"abstract":"\u0000 \u0000 The Mycobacterium tuberculosis complex (MTBC) bacteria include the slowly growing, host-associated bacteria Mycobacterium tuberculosis, Mycobacterium Bovis, Mycobacterium microti, Mycobacterium africanum, Mycobacterium pinnipedii.\u0000 \u0000 \u0000 \u0000 Comparative Functional Genomics Studies for understanding the Hypothetical Proteins in Mycobacterium tuberculosis variant microti 12.\u0000 \u0000 \u0000 \u0000 A computational genomics study was performed to understand the 247 hypothetical protein genes. Functional annotation of virtual proteins was performed on different servers to maximize confidence level.\u0000 \u0000 \u0000 \u0000 Sequence Retrieval. The whole genome sequences for the Mycobacterium tuberculosis micro variant 12 were retrieved from the KEGG database (\u0000 http://www.genome.jp/kegg/) and were used for screening 247 hypothetical proteins (Fig.\u0000 \u0000 1\u0000 ). Functional Annotation and Sub-cellular localization. The Mycobacterium tuberculosis micro variant 12 hypothetical proteins were screened and sorted out from the genome and were individually analyzed for the presence of conserved functional domains by using computational biology tools like CDD-BLAST (\u0000 https://www.ncbi.nlm.nih.gov/Structure/cdd/wrpsb.cgi) ;Pfam (\u0000 http://pfam.xfam.org/ncbiseq/398365647); The subcellular localization of hypothetical proteins was determined by CELLO2GO (\u0000 http://cello.life.nctu.edu.tw). These web tools can search the defined conserved domains in the sequences available in the online servers or databases and assist in the classification of proteins in the appropriate families. Protein Structure Prediction. The\u0000 in-silico structure predictions of the hypothetical protein sequences showing functional properties were carried out by using the PS2 Protein Structure Prediction Server (\u0000 http://www.ps2.life.nctu.edu.tw/). The online server helps to generate the 3D structures of the hypothetical proteins. The server accepts the sequences in FASTA format as a query to generate resultant proteins 3D structures. The structure determination is completely based on the conserved template regions detected during functional annotations. Protein-protein interaction through String database: The interaction of each hypothetical protein analyzed for functional characteristics was subjected to a protein-protein interaction server for the prediction of a possible functional role in interaction amongst the available known proteins (\u0000 https://string-db.org/). This information can help us to further validated the functional role of such hypothetical proteins and their possible role in the Mycobacterium Tuberculosis micro variant. Protein secondary structure prediction through JPred4: The secondary structure prediction of all the hypothetical proteins was determined through JPred4 (\u0000 http://www.compbio.dundee.ac.uk/jpred4/index.html) and served to identify the available secondary structures in the unknown hypothetical protein sequences. These further help us to understand the available templates in the uncharacterized protein seq","PeriodicalId":38956,"journal":{"name":"Open Bioinformatics Journal","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-07-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"47867946","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Integrated Bioinformatics Approach for Disclosing Autophagy Pathway as a Therapeutic Target in Advanced KRAS Mutated/Positive Lung Adenocarcinoma","authors":"Yasmeen I Dodin","doi":"10.2174/18750362-v16-2305230-2022-18","DOIUrl":"https://doi.org/10.2174/18750362-v16-2305230-2022-18","url":null,"abstract":"\u0000 \u0000 Lung cancer is the leading cause of cancer-related deaths, accounting for 1.8 million deaths (18%). Nearly 80%-85% of lung cancer cases are non-small cell lung cancers (NSCLC). One of the most frequent genetic mutations in NSCLC is the Kirsten Rat Sarcoma Oncogene Homolog (KRAS) gene mutation. In recent years, autophagy has drawn substantial attention as a potential pathway that can be targeted in cancer driven by KRAS gene mutation to efficiently improve the therapeutic profile of different treatments.\u0000 \u0000 \u0000 \u0000 In this study, we have investigated the potential of targeting the autophagy pathway as a treatment approach in advanced KRAS-mutated lung adenocarcinoma using gene expression data from The Cancer Genome Atlas Lung Adenocarcinoma (TCGA-LUAD) project.\u0000 \u0000 \u0000 \u0000 Compared to KRAS wild-type lung adenocarcinoma, there were found 11 differentially expressed autophagy-related genes (DEARGs), with 5 upregulated and 6 downregulated DEARGs (threshold of adjusted p-value <0.05).\u0000 \u0000 \u0000 \u0000 These DEARGs can be investigated as potential genes that can be targeted by different autophagy inhibitors.\u0000","PeriodicalId":38956,"journal":{"name":"Open Bioinformatics Journal","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-05-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"46535451","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Computational Analysis, In silico Functional Annotation, and Expression of Recombinant PE_PGRS Protein Biomarkers Found in Mycobacterium tuberculosis","authors":"Avanthi Moodley-Reddy, Thamsanqa E. Chiliza, O. Pooe","doi":"10.2174/18750362-v16-e230306-2022-6","DOIUrl":"https://doi.org/10.2174/18750362-v16-e230306-2022-6","url":null,"abstract":"\u0000 \u0000 Over the years, there have been many advances made within the treatment and diagnosis of Mycobacterium Tuberculosis (Mtb). In recent times, the rise of drug resistance has led to higher mortality rates, specifically in poorer countries. There is an urgent need for novel treatment regimens to work against Mtb. Previous studies have identified a gene family within Mtb, known as PE_PGRS proteins, which has shown potential as a drug target. Functional annotations can assist with identifying the role these proteins may play within Mtb.\u0000 \u0000 \u0000 \u0000 Previous studies indicated PE_PGRS to have potential for further research. The protein biomarkers that showed the most promise were identified as PE_PGRS17, PE_PGRS31, PE_PGRS50, and PEPGRS54. The sequences of these proteins were searched on the Mycobrowser software. Results were designed by entering these sequences into various computational algorithms. PE_PGRS17 showed characteristics of a potential vaccine candidate. Considering this result, expression profiling and purification were conducted on the recombinant PE_PGRS17 Mtb protein biomarker.\u0000 \u0000 \u0000 \u0000 The results were calculated using various online software algorithms. Many characteristics were predicted to understand the stability, localization, and function of these proteins. All the proteins have been estimated to produce an immune response or be involved in the process of immunity. The recombinantPE_PGRS17 protein was chosen to be optimally expressed and purified using E.coli as a host cell. These findings specifically on PE_PGRS17, can be expanded in future scientific studies.\u0000 \u0000 \u0000 \u0000 The predicted structures, protein-protein interaction, and antigenic properties of the proteins estimate whether a protein can be used for further studies, specifically as drug/vaccine targets. Ultimately, PE_PGRS17 is seen as the most stable according to its predicted structure, which holds promise as a key factor in future tuberculosis studies.\u0000","PeriodicalId":38956,"journal":{"name":"Open Bioinformatics Journal","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-04-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"48588996","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"The Development and Progress in Machine Learning for Protein Subcellular Localization Prediction","authors":"Le He, Xiyu Liu","doi":"10.2174/18750362-v15-e2208110","DOIUrl":"https://doi.org/10.2174/18750362-v15-e2208110","url":null,"abstract":"Protein subcellular localization is a novel and promising area and is defined as searching for the specific location of proteins inside the cell, such as in the nucleus, in the cytoplasm or on the cell membrane. With the rapid development of next-generation sequencing technology, more and more new protein sequences have been continuously discovered. It is no longer sufficient to merely use traditional wet experimental methods to predict the subcellular localization of these new proteins. Therefore, it is urgent to develop high-throughput computational methods to achieve quick and precise protein subcellular localization predictions. This review summarizes the development of prediction methods for protein subcellular localization over the past decades, expounds on the application of various machine learning methods in this field, and compares the properties and performance of various well-known predictors. The narrative of this review mainly revolves around three main types of methods, namely, the sequence-based methods, the knowledge-based methods, and the fusion methods. A special focus is on the gene ontology (GO)-based methods and the PLoc series methods. Finally, this review looks forward to the future development directions of protein subcellular localization prediction.","PeriodicalId":38956,"journal":{"name":"Open Bioinformatics Journal","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-10-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"41748283","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Unsupervised Deep learning-based Feature Fusion Approach for Detection and Analysis of COVID-19 using X-ray and CT Images","authors":"Vinayakumar Ravi, T. Pham","doi":"10.2174/18750362-v15-e2207290","DOIUrl":"https://doi.org/10.2174/18750362-v15-e2207290","url":null,"abstract":"\u0000 \u0000 This study investigates an unsupervised deep learning-based feature fusion approach for the detection and analysis of COVID-19 using chest X-ray (CXR) and Computed tomography (CT) images.\u0000 \u0000 \u0000 \u0000 The outbreak of COVID-19 has affected millions of people all around the world and the disease is diagnosed by the reverse transcription-polymerase chain reaction (RT-PCR) test which suffers from a lower viral load, and sampling error, etc. Computed tomography (CT) and chest X-ray (CXR) scans can be examined as most infected people suffer from lungs infection. Both CT and CXR imaging techniques are useful for the COVID-19 diagnosis at an early stage and it is an alternative to the RT-PCR test.\u0000 \u0000 \u0000 \u0000 The manual diagnosis of CT scans and CXR images are labour-intensive and consumes a lot of time. To handle this situation, many AI-based solutions are researched including deep learning-based detection models, which can be used to help the radiologist to make a better diagnosis. However, the availability of annotated data for COVID-19 detection is limited due to the need for domain expertise and expensive annotation cost. Also, most existing state-of-the-art deep learning-based detection models follow a supervised learning approach. Therefore, in this work, we have explored various unsupervised learning models for COVID-19 detection which does not need a labelled dataset.\u0000 \u0000 \u0000 \u0000 In this work, we propose an unsupervised deep learning-based COVID-19 detection approach that incorporates the feature fusion method for performance enhancement. Four different sets of experiments are run on both CT and CXR scan datasets where convolutional autoencoders, pre-trained CNNs, hybrid, and PCA-based models are used for feature extraction and K-means and GMM techniques are used for clustering.\u0000 \u0000 \u0000 \u0000 The maximum accuracy of 84% is achieved by the model Autoencoder3-ResNet50 (GMM) on the CT dataset and for the CXR dataset, both Autoencoder1-VGG16 (KMeans and GMM) models achieved 70% accuracy.\u0000 \u0000 \u0000 \u0000 Our proposed deep unsupervised learning, feature fusion-based COVID-19 detection approach achieved promising results on both datasets. It also outperforms four well-known existing unsupervised approaches.\u0000","PeriodicalId":38956,"journal":{"name":"Open Bioinformatics Journal","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-09-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"41661571","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}