Research on laparoscopic surgical instrument detection technology based on multi-attention enhanced feature pyramid network

Research on laparoscopic surgical instrument detection technology based on multi-attention enhanced feature pyramid 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 Research Article Research on laparoscopic surgical instrument detection technology based on multi-attention enhanced feature pyramid network Xinying Wang, Yuxuan Zhang, Yang Li This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-1730847/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 8 You are reading this latest preprint version Abstract Laparoscopic surgical instruments are difficult in object detection due to their large-scale changes. Feature Pyramid Network (FPN) can effectively solve the multi-scale object detection problem. However, there are still some problems in FPN that limit the full utilization of multi-scale features. By analyzing the design problem of FPN, we propose the Multi-Attention Enhanced Feature Pyramid Network (MAFPN). First, we replace the convolutional block with feature selection module (FSM) that combines channel attention and global attention, which selectively maintains important information and enhances the expressiveness of features at each scale. Second, the global contextual information is captured by the self-attentive augmented fusion module (AAFM), which enriches the high-level feature information in the FPN and enhances the feature fusion effect. Finally, we use Dynamic Convolution Decomposition (DCD) to alleviate the impact of upsampling while enhancing the feature expression ability. Experimental results on the laparoscopic surgical instrument detection dataset m2cai16-tools-locations indicate that when ResNet50 is used as the backbone network, MAFPN improves the baseline network by 1.8 percentage points, to 96.5AP. Laparoscopic Surgical Instrument Detection Feature Pyramid Attention mechanism dynamic convolution Object Detection Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Review Version 1 posted Editorial decision: Major revision 10 Aug, 2022 Reviews received at journal 29 Jul, 2022 Reviewers agreed at journal 23 Jul, 2022 Reviewers agreed at journal 19 Jul, 2022 Reviewers invited by journal 18 Jul, 2022 Editor assigned by journal 14 Jul, 2022 Submission checks completed at journal 09 Jun, 2022 First submitted to journal 06 Jun, 2022 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-1730847","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":112402781,"identity":"5608264d-9d0a-4551-9b9f-b74f8a979034","order_by":0,"name":"Xinying Wang","email":"","orcid":"","institution":"Changchun University of Technology","correspondingAuthor":false,"prefix":"","firstName":"Xinying","middleName":"","lastName":"Wang","suffix":""},{"id":112402782,"identity":"7519bcbd-b0f6-4f64-9ab9-d5928d1e43a8","order_by":1,"name":"Yuxuan Zhang","email":"","orcid":"","institution":"Changchun University of Technology","correspondingAuthor":false,"prefix":"","firstName":"Yuxuan","middleName":"","lastName":"Zhang","suffix":""},{"id":112402783,"identity":"3cad9339-a0a0-48b9-b150-1ce6fd1dfcf0","order_by":2,"name":"Yang Li","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABCUlEQVRIiWNgGAWjYBACPmY4k7Hx8Z8KGwY2QlrYEFqYmw14zqQRoQXBZG+T4G07TNhhbOw8hp8Lfh2WNzh+sE1Csu28PZ908wOGHxXb8DiMx1h6Zt9hww1nEpstDM7dTmyTOWbA2HPmNj4tBtK8PYcZNxxIbLyRUHY7gU0iwYCZsQ2vFuPfQC32G84/bJA4wHbOnk0i/QMhLWbSPD8OJ264kdgk2dB2gLFNIoeQLWxl1rwN6ckzbzxsNmY4k5wI1FJwEJ9f+PkPb77N88fatu98+sPHDBV29vIz0jc++FGBWwsDA4cBA2NbM4PCASSxAzjUQgH7AwaGP3UM8g34lY2CUTAKRsEIBgDhvFj2wHMQSQAAAABJRU5ErkJggg==","orcid":"","institution":"Changchun University of Technology","correspondingAuthor":true,"prefix":"","firstName":"Yang","middleName":"","lastName":"Li","suffix":""}],"badges":[],"createdAt":"2022-06-06 15:29:22","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-1730847/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-1730847/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":22688637,"identity":"b67d9551-96c7-4348-84be-a534acb44cd6","added_by":"auto","created_at":"2022-06-15 16:28:38","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":756565,"visible":true,"origin":"","legend":"","description":"","filename":"Researchonlaparoscopicsurgicalinstrumentdetectiontechnologybasedonmultiattentionenhancedfeaturepyramidnetwork.pdf","url":"https://assets-eu.researchsquare.com/files/rs-1730847/v1_covered.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Research on laparoscopic surgical instrument detection technology based on multi-attention enhanced feature pyramid network","fulltext":[{"header":"Full Text","content":"This preprint is available for \u003ca href='/article/rs-1730847/latest.pdf' target='_blank'\u003edownload as a PDF\u003c/a\u003e."}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":true,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"signal-image-and-video-processing","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"sivp","sideBox":"Learn more about [Signal, Image and Video Processing](http://link.springer.com/journal/11760)","snPcode":"11760","submissionUrl":"https://submission.nature.com/new-submission/11760/3","title":"Signal, Image and Video Processing","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"Laparoscopic Surgical Instrument Detection, Feature Pyramid, Attention mechanism, dynamic convolution Object Detection","lastPublishedDoi":"10.21203/rs.3.rs-1730847/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-1730847/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eLaparoscopic surgical instruments are difficult in object detection due to their large-scale changes. Feature Pyramid Network (FPN) can effectively solve the multi-scale object detection problem. However, there are still some problems in FPN that limit the full utilization of multi-scale features. By analyzing the design problem of FPN, we propose the Multi-Attention Enhanced Feature Pyramid Network (MAFPN). First, we replace the convolutional block with feature selection module (FSM) that combines channel attention and global attention, which selectively maintains important information and enhances the expressiveness of features at each scale. Second, the global contextual information is captured by the self-attentive augmented fusion module (AAFM), which enriches the high-level feature information in the FPN and enhances the feature fusion effect. Finally, we use Dynamic Convolution Decomposition (DCD) to alleviate the impact of upsampling while enhancing the feature expression ability. Experimental results on the laparoscopic surgical instrument detection dataset m2cai16-tools-locations indicate that when ResNet50 is used as the backbone network, MAFPN improves the baseline network by 1.8 percentage points, to 96.5AP.\u0026nbsp;\u003c/p\u003e","manuscriptTitle":"Research on laparoscopic surgical instrument detection technology based on multi-attention enhanced feature pyramid network","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2022-06-15 16:28:31","doi":"10.21203/rs.3.rs-1730847/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Major revision","date":"2022-08-10T11:36:57+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2022-07-29T22:49:17+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"0e2c1fc0-dba0-4705-a22a-d0e844fea83a","date":"2022-07-23T13:29:01+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"386f6c2d-56df-44af-bcdb-ae904c6fdfe6","date":"2022-07-19T18:11:37+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2022-07-18T15:34:25+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2022-07-14T12:34:29+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2022-06-09T15:31:02+00:00","index":"","fulltext":""},{"type":"submitted","content":"Signal, Image and Video Processing","date":"2022-06-06T15:25:50+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"signal-image-and-video-processing","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"sivp","sideBox":"Learn more about [Signal, Image and Video Processing](http://link.springer.com/journal/11760)","snPcode":"11760","submissionUrl":"https://submission.nature.com/new-submission/11760/3","title":"Signal, Image and Video Processing","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"cddf1dad-d87b-46cc-8184-ec30594aac95","owner":[],"postedDate":"June 15th, 2022","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2022-12-04T20:14:14+00:00","versionOfRecord":[],"versionCreatedAt":"2022-06-15 16:28:31","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-1730847","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-1730847","identity":"rs-1730847","version":["v1"]},"buildId":"J0_U0BvcaRcwD8yVFaRlm","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}