Radiographic Pneumonia Detection and Multiclass Classification Using Deep Learning Models

Radiographic Pneumonia Detection and Multiclass Classification Using Deep Learning Models | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Radiographic Pneumonia Detection and Multiclass Classification Using Deep Learning Models Ayenew Walle Kebede, Melesew Mossie Beyene, Addisu Taye Tamene, and 1 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8573233/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract The high prevalence of pneumonia in Ethiopia, where diagnostic tools are scarce and access to healthcare is uneven, particularly in rural areas, is addressed in this paper by presenting an advanced method for automating pneumonia detection and classification using deep learning and image processing techniques. Rapid, precise and automated diagnoses are provided by the suggested method, which enhances healthcare delivery and lowers mortality. The method starts with systematic preprocessing of chest X-ray images, which includes contrast enhancement, noise reduction and normalization to guarantee high-quality inputs for analysis. To increase the performance of Convolutional Neural Networks (CNNs), the study uses Gabor filters for feature extraction, which enhances textural information. The images are categorized into three classes using a three-way SoftMax classifier normal, viral and bacterial pneumonia. The best-performing CNN model out of the four that were tested was DenseNet201 with a testing accuracy of 97.39% and a training accuracy of 99.40%. This high level of accuracy indicates that the proposed system is highly effective in distinguishing between normal and pathological lung conditions, suggesting its potential for enhancing diagnostic precision and efficiency in clinical settings. Pneumonia Deep learning CNN Gabor filter X-ray images Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-8573233","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":584295783,"identity":"51d0399a-92f1-4578-8940-17ea8553a617","order_by":0,"name":"Ayenew Walle Kebede","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA30lEQVRIiWNgGAWjYBACNjBZYMHAJv/4AJAlIUOkFgMJBn6GtASQFh4i7QJqkWzIMQAxCWvhk0hgk/hhICFncODM51c3aix4GNgPH92A12FALZI9BhLGBgd7t1nnHAM6jCct7QYhLTd4DCQSNxzm3WacwwbUIsFjRlDLzT8GEvUbjvE8M875R6SW20BbEiR7eJgf57YRo4XnYftvGQMJw34JNjPm3D4JHjZCfpFvTz5s+KbCRp5Ngvnx55xvdXL87IeP4dXCwMDYgHAkmMSvHBUwfyBF9SgYBaNgFIwcAAD/kz07hVVycwAAAABJRU5ErkJggg==","orcid":"","institution":"","correspondingAuthor":true,"prefix":"","firstName":"Ayenew","middleName":"Walle","lastName":"Kebede","suffix":""},{"id":584295784,"identity":"be83c6f3-f969-4107-80e8-808f75dcce5a","order_by":1,"name":"Melesew Mossie Beyene","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Melesew","middleName":"Mossie","lastName":"Beyene","suffix":""},{"id":584295786,"identity":"d137dd32-0926-4e1f-82fd-fb5de40c318d","order_by":2,"name":"Addisu Taye Tamene","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Addisu","middleName":"Taye","lastName":"Tamene","suffix":""},{"id":584295787,"identity":"2cb7af87-8e13-4e4c-9592-8600cdce3a6d","order_by":3,"name":"Amare Gedif Yalew","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Amare","middleName":"Gedif","lastName":"Yalew","suffix":""}],"badges":[],"createdAt":"2026-01-11 11:38:13","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8573233/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8573233/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":103529392,"identity":"c1bacd3c-34d3-4962-9f53-3048bde4aeee","added_by":"auto","created_at":"2026-02-26 16:40:34","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1001763,"visible":true,"origin":"","legend":"","description":"","filename":"Manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8573233/v1_covered_701c8a5b-3fd5-45b0-ac7f-6b4d00bdadc2.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Radiographic Pneumonia Detection and Multiclass Classification Using Deep Learning Models","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":true,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":true,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Pneumonia, Deep learning, CNN, Gabor filter, X-ray images","lastPublishedDoi":"10.21203/rs.3.rs-8573233/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8573233/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eThe high prevalence of pneumonia in Ethiopia, where diagnostic tools are scarce and access to healthcare is uneven, particularly in rural areas, is addressed in this paper by presenting an advanced method for automating pneumonia detection and classification using deep learning and image processing techniques. Rapid, precise and automated diagnoses are provided by the suggested method, which enhances healthcare delivery and lowers mortality. The method starts with systematic preprocessing of chest X-ray images, which includes contrast enhancement, noise reduction and normalization to guarantee high-quality inputs for analysis. To increase the performance of Convolutional Neural Networks (CNNs), the study uses Gabor filters for feature extraction, which enhances textural information. The images are categorized into three classes using a three-way SoftMax classifier normal, viral and bacterial pneumonia. The best-performing CNN model out of the four that were tested was DenseNet201 with a testing accuracy of 97.39% and a training accuracy of 99.40%. This high level of accuracy indicates that the proposed system is highly effective in distinguishing between normal and pathological lung conditions, suggesting its potential for enhancing diagnostic precision and efficiency in clinical settings.\u003c/p\u003e","manuscriptTitle":"Radiographic Pneumonia Detection and Multiclass Classification Using Deep Learning Models","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-02-03 17:31:10","doi":"10.21203/rs.3.rs-8573233/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"7a443533-75f9-4898-b073-1869a5794b56","owner":[],"postedDate":"February 3rd, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2026-02-26T16:40:00+00:00","versionOfRecord":[],"versionCreatedAt":"2026-02-03 17:31:10","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8573233","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8573233","identity":"rs-8573233","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}