Segmentation of Non-Small Cell Lung Carcinomas: Introducing DRU-Net and Multi-Lens Distortion

Segmentation of Non-Small Cell Lung Carcinomas: Introducing DRU-Net and Multi-Lens Distortion | 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 Segmentation of Non-Small Cell Lung Carcinomas: Introducing DRU-Net and Multi-Lens Distortion Soroush Oskouei, Marit Valla, André Pedersen, Erik Smistad, Vibeke Grotnes Dale, and 8 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-4601446/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 Considering the increased workload in pathology laboratories today, automated tools such as artificial intelligence models can help pathologists with their tasks and ease the workload. In this paper, we are proposing a segmentation model (DRU-Net) that can provide a delineation of human non-small cell lung carcinomas and an augmentation method that can improve classification results. The proposed model is a fused combination of truncated pre-trained DenseNet201 and ResNet101V2 as a patch-wise classifier followed by a lightweight U-Net as a refinement model. We have used two datasets (Norwegian Lung Cancer Biobank and Haukeland University Hospital lung cancer cohort) to create our proposed model. The DRU-Net model achieves an average of 0.91 Dice similarity coefficient. The proposed spatial augmentation method (multi-lens distortion) improved the network performance by 3%. Our findings show that choosing image patches that specifically include regions of interest leads to better results for the patch-wise classifier compared to other sampling methods. The qualitative analysis showed that the DRU-Net model is generally successful in detecting the tumor. On the test set, some of the cases showed areas of false positive and false negative segmentation in the periphery, particularly in tumors with inflammatory and reactive changes. Biological sciences/Cancer/Lung cancer Physical sciences/Mathematics and computing/Computer science Full Text Additional Declarations No competing interests reported. Supplementary Files SupplementaryInformationforScientificReports.zip SupplementaryFigureS1.png SupplementaryFigureS2.png 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-4601446","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":335510697,"identity":"f9f5de8f-3b58-4c22-a649-9dc7fc01e3aa","order_by":0,"name":"Soroush Oskouei","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA4UlEQVRIiWNgGAWjYLACxgYgwQNiGdiQriUNQpOgheEwYS3y7b3PJH/uYJCT7zl8TOJDwfk8e+keA+aCCtxaDM4cN5PmPcNgbHC2LU1yhsHtYh6ZMwbMM87g0SKRxibN2MaQuIGfx0yax+B2Yo9EjgEzbxseh81IY5P8CdQyvx+o5Y/BOcJaGG6ksUkAFSQ2nO0xk2YwOEBYi8GZY8zWvG0SxkBGsmWPQXJiz420gsM8ePwi397GePNnmw0wxJIP3vjxxy6xfUbyxsc8eEIMCiRQuQcIahgFo2AUjIJRgBcAACdDRukx03z+AAAAAElFTkSuQmCC","orcid":"","institution":"Norwegian University of Science and Technology (NTNU), Department of Circulation and Medical Imaging","correspondingAuthor":true,"prefix":"","firstName":"Soroush","middleName":"","lastName":"Oskouei","suffix":""},{"id":335510698,"identity":"aa911892-e96d-44cb-bb32-ef901815f425","order_by":1,"name":"Marit Valla","email":"","orcid":"","institution":"Norwegian University of Science and Technology (NTNU), Department of Clinical and Molecular Medicine","correspondingAuthor":false,"prefix":"","firstName":"Marit","middleName":"","lastName":"Valla","suffix":""},{"id":335510699,"identity":"2b0b0b67-e2f5-46d0-9430-ff8ba000e56c","order_by":2,"name":"André Pedersen","email":"","orcid":"","institution":"Norwegian University of Science and Technology (NTNU), Department of Clinical and Molecular Medicine","correspondingAuthor":false,"prefix":"","firstName":"André","middleName":"","lastName":"Pedersen","suffix":""},{"id":335510700,"identity":"5acdef30-b03a-4b85-b266-4e80b4217d01","order_by":3,"name":"Erik Smistad","email":"","orcid":"","institution":"Norwegian University of Science and Technology (NTNU), Department of Circulation and Medical Imaging","correspondingAuthor":false,"prefix":"","firstName":"Erik","middleName":"","lastName":"Smistad","suffix":""},{"id":335510701,"identity":"373c8d41-259e-4267-bde4-768c363979fd","order_by":4,"name":"Vibeke Grotnes Dale","email":"","orcid":"","institution":"Norwegian University of Science and Technology (NTNU), Department of Clinical and Molecular Medicine","correspondingAuthor":false,"prefix":"","firstName":"Vibeke","middleName":"Grotnes","lastName":"Dale","suffix":""},{"id":335510702,"identity":"bb5c1bfe-de7c-4be7-9356-a8e4461f7e05","order_by":5,"name":"Maren Høibø","email":"","orcid":"","institution":"Norwegian University of Science and Technology (NTNU), Department of Clinical and Molecular Medicine","correspondingAuthor":false,"prefix":"","firstName":"Maren","middleName":"","lastName":"Høibø","suffix":""},{"id":335510703,"identity":"373fd1e0-6aad-4f50-a295-28c87b680ce7","order_by":6,"name":"Sissel Gyrid Freim Wahl","email":"","orcid":"","institution":"St. Olavs hospital, Trondheim University Hospital, Department of Pathology","correspondingAuthor":false,"prefix":"","firstName":"Sissel","middleName":"Gyrid Freim","lastName":"Wahl","suffix":""},{"id":335510705,"identity":"794cb43b-6be6-4288-99df-da2beef794e8","order_by":7,"name":"Mats Dehli Haugum","email":"","orcid":"","institution":"St. Olavs hospital, Trondheim University Hospital, Department of Pathology","correspondingAuthor":false,"prefix":"","firstName":"Mats","middleName":"Dehli","lastName":"Haugum","suffix":""},{"id":335510706,"identity":"f1620ad4-ec8f-4566-b246-dcd959c7f14a","order_by":8,"name":"Thomas Langø","email":"","orcid":"","institution":"SINTEF Digital, Department of Health Research","correspondingAuthor":false,"prefix":"","firstName":"Thomas","middleName":"","lastName":"Langø","suffix":""},{"id":335510708,"identity":"60716165-d01c-41f4-a26d-4563d8a382c2","order_by":9,"name":"Maria Paula Ramnefjell","email":"","orcid":"","institution":"University of Bergen, Department of Clinical Medicine Centre for Cancer Biomarkers (CCBIO)","correspondingAuthor":false,"prefix":"","firstName":"Maria","middleName":"Paula","lastName":"Ramnefjell","suffix":""},{"id":335510710,"identity":"7448ffb4-d8a2-4b59-85a5-7e7a99cfba4f","order_by":10,"name":"Lars Andreas Akslen","email":"","orcid":"","institution":"University of Bergen, Department of Clinical Medicine Centre for Cancer Biomarkers (CCBIO)","correspondingAuthor":false,"prefix":"","firstName":"Lars","middleName":"Andreas","lastName":"Akslen","suffix":""},{"id":335510711,"identity":"fa54c37c-6e97-4cd8-aec8-adc6a355de03","order_by":11,"name":"Gabriel Kiss","email":"","orcid":"","institution":"St. Olavs hospital, Trondheim University Hospital, Research Department, Center for Medical equipment, Technology, and Innovation","correspondingAuthor":false,"prefix":"","firstName":"Gabriel","middleName":"","lastName":"Kiss","suffix":""},{"id":335510712,"identity":"65d110a4-304d-49a9-8b53-fe8e6dacf26e","order_by":12,"name":"Hanne Sorger","email":"","orcid":"","institution":"Norwegian University of Science and Technology (NTNU), Department of Circulation and Medical Imaging","correspondingAuthor":false,"prefix":"","firstName":"Hanne","middleName":"","lastName":"Sorger","suffix":""}],"badges":[],"createdAt":"2024-06-18 17:06:17","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-4601446/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-4601446/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":77681520,"identity":"2536a7ce-e925-4295-afbc-50bbb7b2d943","added_by":"auto","created_at":"2025-03-04 08:39:25","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":970415,"visible":true,"origin":"","legend":"","description":"","filename":"TemplateforsubmissionstoScientificReports7.pdf","url":"https://assets-eu.researchsquare.com/files/rs-4601446/v1_covered_d7d7df67-3341-4053-a7fb-2378c0579a10.pdf"},{"id":61921575,"identity":"cda1475c-5490-4f6a-97fa-7f179a3dc04c","added_by":"auto","created_at":"2024-08-07 06:15:41","extension":"zip","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":3054040,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryInformationforScientificReports.zip","url":"https://assets-eu.researchsquare.com/files/rs-4601446/v1/e5eb840826ae174f4ec908dc.zip"},{"id":61921573,"identity":"d7ee2d53-d604-4679-8f93-6946dd7647d1","added_by":"auto","created_at":"2024-08-07 06:15:41","extension":"png","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":690364,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryFigureS1.png","url":"https://assets-eu.researchsquare.com/files/rs-4601446/v1/72dbbb26fae520a8eba35c9d.png"},{"id":61921574,"identity":"cdf6f904-aa72-4b5c-a5d4-a091ca6ee47c","added_by":"auto","created_at":"2024-08-07 06:15:41","extension":"png","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":2368982,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryFigureS2.png","url":"https://assets-eu.researchsquare.com/files/rs-4601446/v1/3c18032366073664417816dc.png"}],"financialInterests":"No competing interests reported.","formattedTitle":"Segmentation of Non-Small Cell Lung Carcinomas: Introducing DRU-Net and Multi-Lens Distortion","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":"","lastPublishedDoi":"10.21203/rs.3.rs-4601446/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-4601446/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Considering the increased workload in pathology laboratories today, automated tools such as artificial intelligence models can help pathologists with their tasks and ease the workload. In this paper, we are proposing a segmentation model (DRU-Net) that can provide a delineation of human non-small cell lung carcinomas and an augmentation method that can improve classification results. The proposed model is a fused combination of truncated pre-trained DenseNet201 and ResNet101V2 as a patch-wise classifier followed by a lightweight U-Net as a refinement model. We have used two datasets (Norwegian Lung Cancer Biobank and Haukeland University Hospital lung cancer cohort) to create our proposed model. The DRU-Net model achieves an average of 0.91 Dice similarity coefficient. The proposed spatial augmentation method (multi-lens distortion) improved the network performance by 3%. Our findings show that choosing image patches that specifically include regions of interest leads to better results for the patch-wise classifier compared to other sampling methods. The qualitative analysis showed that the DRU-Net model is generally successful in detecting the tumor. On the test set, some of the cases showed areas of false positive and false negative segmentation in the periphery, particularly in tumors with inflammatory and reactive changes.","manuscriptTitle":"Segmentation of Non-Small Cell Lung Carcinomas: Introducing DRU-Net and Multi-Lens Distortion","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-08-07 06:15:36","doi":"10.21203/rs.3.rs-4601446/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":"d851b61d-0860-4bd4-82e0-f3b0a05e4024","owner":[],"postedDate":"August 7th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":35534798,"name":"Biological sciences/Cancer/Lung cancer"},{"id":35534799,"name":"Physical sciences/Mathematics and computing/Computer science"}],"tags":[],"updatedAt":"2025-03-04T08:38:59+00:00","versionOfRecord":[],"versionCreatedAt":"2024-08-07 06:15:36","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-4601446","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-4601446","identity":"rs-4601446","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}