Detection of Precancerous Cervical Cancer Using Dual-Source Image Fusion Network

Detection of Precancerous Cervical Cancer Using Dual-Source Image Fusion Network | 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 Article Detection of Precancerous Cervical Cancer Using Dual-Source Image Fusion Network Zhang Qianlong, Zeng Yue This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6642645/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 White light medical images are often used to detect early symptoms of cervical cancer, but they have limitations in the diagnosis and grading of early symptoms of cervical cancer. White light images can only provide limited information and it is difficult to fully capture the complex characteristics of cervical cancer,such as metabolic changes and microenvironmental differences. Therefore, it is difficult to judge the severity of symptoms and it is easy to lead to misdiagnosis. This paper proposes a dual-source medical image fusion network PFM (PConvFusionModel) based on attention mechanism for the diagnosis and classification of early cervical cancer symptoms. The network processes two image sources, white light images and fluorescence images, and extracts the morphological features of white light images and the tissue and metabolic features of fluorescence images respectively. PFM is designed based on PConv architecture and includes four core components: feature extraction module, attention enhancement module, dual-source feature fusion module and multi-head self-attention decision module. The network first obtains the original features of the two images through the PConv feature extraction module, then uses the channel attention and spatial attention modules to enhance the features of the two types of images respectively, then splices the enhanced features, and further integrates the relationship between the features through the multi-head self-attention mechanism. Finally, the classification module completes the diagnosis and classification of the disease. The experimental results show that compared with the single-source image neural network method, the accuracy of this method on the experimental data set is improved by 7 percentage points, and the recall rate is improved by 6 percentage points. The experiment verifies that this method improves the accuracy of detecting early symptoms of cervical cancer. Biological sciences/Cancer Health sciences/Oncology Physical sciences/Mathematics and computing Dual-source images attention mechanism image fusion 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-6642645","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":466893921,"identity":"7dad78c6-6bdd-4434-996e-87f5efdd0338","order_by":0,"name":"Zhang Qianlong","email":"","orcid":"","institution":"Jinling Institute of Technology","correspondingAuthor":false,"prefix":"","firstName":"Zhang","middleName":"","lastName":"Qianlong","suffix":""},{"id":466893927,"identity":"81d829fd-49ec-4a43-91bb-bd7ac413fe89","order_by":1,"name":"Zeng Yue","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAvElEQVRIiWNgGAWjYNCCCjYeECVBgpYzbDw8pGlhbGNgIF6LfHvv4de88/hk7BmYD97mYbDLI2xBz7k0a95tIIexJVvzMCQXE9TCLJFjZgzRwmMmzcNwILGBkBY2+TdALXNAWvi/EaeFR4LH+DFvA9gWNuK0SPDkmDHOOQbUcpjN2HKOQTJhLfLtZ4w/vKk5Zs/e3vzwxpsKO8JaQN6R4mE4BgwIENuACPVAwPzxB0MNcUpHwSgYBaNgZAIA4MEtG0BfScgAAAAASUVORK5CYII=","orcid":"","institution":"Jinling Institute of Technology","correspondingAuthor":true,"prefix":"","firstName":"Zeng","middleName":"","lastName":"Yue","suffix":""}],"badges":[],"createdAt":"2025-05-12 04:08:23","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6642645/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6642645/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":86320887,"identity":"31f3dfb2-f4bf-4dbe-83a1-56d0f217f17e","added_by":"auto","created_at":"2025-07-09 09:39:26","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1011908,"visible":true,"origin":"","legend":"","description":"","filename":"Researchonearlycervicalcancerdetectionmethodbasedondualsourceimagefusionnetwork.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6642645/v1_covered_d5becb9a-01ad-4203-b03a-92e14beb99f1.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Detection of Precancerous Cervical Cancer Using Dual-Source Image Fusion Network","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":"Dual-source images, attention mechanism, image fusion","lastPublishedDoi":"10.21203/rs.3.rs-6642645/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6642645/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eWhite light medical images are often used to detect early symptoms of cervical cancer, but they have limitations in the diagnosis and grading of early symptoms of cervical cancer. White light images can only provide limited information and it is difficult to fully capture the\u003c/p\u003e \u003cp\u003ecomplex characteristics of cervical cancer,such as metabolic changes and microenvironmental differences. Therefore, it is difficult to judge the severity of symptoms and it is easy to lead to misdiagnosis. This paper proposes a dual-source medical image fusion network PFM (PConvFusionModel) based on attention mechanism for the diagnosis and classification of early cervical cancer symptoms. The network processes two image sources, white light images and fluorescence images, and extracts the morphological features of white light images and the tissue and metabolic features of fluorescence images respectively. PFM is designed based on PConv architecture and includes four core components: feature extraction module, attention enhancement module, dual-source feature fusion module and multi-head self-attention decision module. The network first obtains the original features of the two images through the PConv feature extraction module, then uses the channel attention and spatial attention modules to enhance the features of the two types of images respectively, then splices the enhanced features, and further integrates the relationship between the features through the multi-head self-attention mechanism. Finally, the classification module completes the diagnosis and classification of the disease. The experimental results show that compared with the single-source image neural network method, the accuracy of this method on the experimental data set is improved by 7 percentage points, and the recall rate is improved by 6 percentage points. The experiment verifies that this method improves the accuracy of detecting early symptoms of cervical cancer.\u003c/p\u003e","manuscriptTitle":"Detection of Precancerous Cervical Cancer Using Dual-Source Image Fusion Network","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-06-06 16:38:34","doi":"10.21203/rs.3.rs-6642645/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":"651522c9-428c-4d0a-b25b-f98978f4c486","owner":[],"postedDate":"June 6th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":49569743,"name":"Biological sciences/Cancer"},{"id":49569744,"name":"Health sciences/Oncology"},{"id":49569745,"name":"Physical sciences/Mathematics and computing"}],"tags":[],"updatedAt":"2025-07-09T09:39:10+00:00","versionOfRecord":[],"versionCreatedAt":"2025-06-06 16:38:34","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-6642645","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6642645","identity":"rs-6642645","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}