Identification of Coronary Calcifications in Optical Coherence Tomography Imaging Using Deep Learning

Identification of Coronary Calcifications in Optical Coherence Tomography Imaging Using Deep Learning | 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 Identification of Coronary Calcifications in Optical Coherence Tomography Imaging Using Deep Learning Yarden Avital, Akiva Madar, Shlomi Arnon, Edward Koifman This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-126929/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 10 You are reading this latest preprint version Abstract Coronary calcifications are an obstacle for successful percutaneous treatment of coronary artery disease patients. The optimal method for delineating calcifications extent is optical coherence tomography (coronary OCT). To identify calcification on OCT and subsequently tailor the appropriate treatment, requires expertise in both image acquisition and interpretation. Image acquisition consists from system calibration, blood clearance by a contrast agent along with synchronization of the pullback process. Accurate interpretation demands careful review by the operator of a segment of 50-75mm of the coronary vessel at steps of 0.5-1mm accounting for 75-100 images in each OCT run, which is time consuming and necessitates some expertise in OCT analysis. In this paper we developed a new deep learning algorithm to assist the physician to identify and quantify coronary calcifications promptly, efficiently and accurately. Our algorithm achieves an accuracy of 0.9903 ± 0.009 over the test set at size of 1500 frames and even managed to find calcifications that weren’t recognized manually by the physician. For the best knowledge of the authors our algorithm achieves high accuracy which was never achieved in the past. Nuclear Medicine & Medical Imaging Bioinformatics Cardiac & Cardiovascular Systems Computational Biology Optical coherence tomography Coronary calcifications Deep Learning Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Full Text Cite Share Download PDF Status: Under Review Version 1 posted Editorial decision: Major revision 25 Feb, 2021 Reviews received at journal 22 Feb, 2021 Reviews received at journal 22 Feb, 2021 Reviewers agreed at journal 14 Feb, 2021 Reviewers agreed at journal 13 Feb, 2021 Reviewers invited by journal 24 Dec, 2020 Editor assigned by journal 17 Dec, 2020 Editor invited by journal 17 Dec, 2020 Submission checks completed at journal 17 Dec, 2020 First submitted to journal 11 Dec, 2020 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-126929","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":6631847,"identity":"5e6dae69-baa6-4eb2-a15c-c0c2cb3127c4","order_by":0,"name":"Yarden Avital","email":"","orcid":"","institution":"Ben-Gurion University of the Negev","correspondingAuthor":false,"prefix":"","firstName":"Yarden","middleName":"","lastName":"Avital","suffix":""},{"id":6631848,"identity":"6c3a7862-a5f8-4ed0-8c1f-e27bc6a0cba6","order_by":1,"name":"Akiva Madar","email":"","orcid":"","institution":"Ben-Gurion University of the Negev","correspondingAuthor":false,"prefix":"","firstName":"Akiva","middleName":"","lastName":"Madar","suffix":""},{"id":6631849,"identity":"4defd6c6-9a3d-436c-8a61-2da18f18204c","order_by":2,"name":"Shlomi Arnon","email":"","orcid":"","institution":"Ben-Gurion University of the Negev","correspondingAuthor":false,"prefix":"","firstName":"Shlomi","middleName":"","lastName":"Arnon","suffix":""},{"id":6631850,"identity":"4a2b199a-d820-4dd8-88b0-800b5ceee477","order_by":3,"name":"Edward Koifman","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA4UlEQVRIiWNgGAWjYFCCBIMDDAdADGYQKSFDiha2BJAWHqK0MEC08BiASYIa+NmTNx74cMYmmn92z+dXN2oseBjYDx/dgE+LZM+zgoMzbqTlzrhzdpt1zjGgw3jS0m7g02JwI8fgMM+Hw7kNN3K3GeewAbVI8Jjh1WIP0vIHqGX+jZxnxjn/iNBiIAHUwnDjcO6GGznMj3PbiNAicQbol54zabkbb6SZMef2SfCwEfILf3vy5g8/jtnkzruR/Phzzrc6OX72w8fwakEGbBJgkljlIMD8gRTVo2AUjIJRMHIAAGbPUyLjnCVwAAAAAElFTkSuQmCC","orcid":"","institution":"Ben-Gurion University of the Negev","correspondingAuthor":true,"prefix":"","firstName":"Edward","middleName":"","lastName":"Koifman","suffix":""}],"badges":[],"createdAt":"2020-12-11 20:14:03","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-126929/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-126929/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":4427318,"identity":"2b25450e-707b-4d6b-a519-949adabc9186","added_by":"auto","created_at":"2020-12-21 22:04:55","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":111780,"visible":true,"origin":"","legend":"Identification of calcified segments flow chart. (Envelope of the artery – vessel wall characteristics, symmetricity, irregularities etc.)","description":"","filename":"Fig1.png","url":"https://assets-eu.researchsquare.com/files/rs-126929/v1/42d18a92d375a08613dcee79.png"},{"id":4427148,"identity":"a954ac59-6f6c-4b4e-a3b7-1fc06fe7f64c","added_by":"auto","created_at":"2020-12-21 21:58:55","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":590095,"visible":true,"origin":"","legend":"Example of single frame in the data set we created. The calcified area in the coronary is marked with green in the annotated image.","description":"","filename":"Fig2.png","url":"https://assets-eu.researchsquare.com/files/rs-126929/v1/06c4b21c3c7428fdf0695f03.png"},{"id":4427222,"identity":"f69f4023-dff3-44de-a0bd-213324810b52","added_by":"auto","created_at":"2020-12-21 22:01:55","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":58415,"visible":true,"origin":"","legend":"Visual block description. blue arrow stands for convolution layer operation, red arrow stands for Max pooling operation, green arrow stands for up-sampling with zero padding and blue boxes describe the size of the feature maps.","description":"","filename":"Fig3.png","url":"https://assets-eu.researchsquare.com/files/rs-126929/v1/05e75863261bdc5222e44d46.png"},{"id":4427225,"identity":"eef71e01-16d6-4552-9d60-502cf3c9272e","added_by":"auto","created_at":"2020-12-21 22:01:55","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":73522,"visible":true,"origin":"","legend":"Deep learning Model architecture described by the fundamental blocks.","description":"","filename":"Fig4.png","url":"https://assets-eu.researchsquare.com/files/rs-126929/v1/c79ac18716355dbc26b13a29.png"},{"id":4427152,"identity":"0d40d7f4-5740-477c-958f-0f6497b6dcca","added_by":"auto","created_at":"2020-12-21 21:58:55","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":661990,"visible":true,"origin":"","legend":"Visualize Results. Two examples of automated segmentation. Legend: A) High accuracy of the model B) Improved accuracy over manual annotation. 1) Original Image. 2) Manual annotation. 3) Automatic segmentation.","description":"","filename":"Fig5.png","url":"https://assets-eu.researchsquare.com/files/rs-126929/v1/e8aa00aad56be1190b2b0d5e.png"},{"id":4427224,"identity":"bb135b31-5649-408f-b06c-08d6da30a89f","added_by":"auto","created_at":"2020-12-21 22:01:55","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":93189,"visible":true,"origin":"","legend":"These graphs describe the accuracy and loss values throughout the learning process. x-axis show the number of epochs, y-axis show Accuracy and loss values respectively.","description":"","filename":"Fig6.png","url":"https://assets-eu.researchsquare.com/files/rs-126929/v1/08f4e16c9b65e1cd8c1be7f6.png"},{"id":13567085,"identity":"a47d1689-9b11-4a2d-9f21-fec444e5814d","added_by":"auto","created_at":"2021-09-17 03:31:31","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":726122,"visible":true,"origin":"","legend":"","description":"","filename":"OCT13092020.pdf","url":"https://assets-eu.researchsquare.com/files/rs-126929/v1_covered.pdf"},{"id":4427411,"identity":"a7132dfa-16d9-4cfc-8711-3568493121fd","added_by":"auto","created_at":"2020-12-21 22:08:00","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":652534,"visible":true,"origin":"","legend":"","description":"","filename":"OCT13092020.pdf","url":"https://assets-eu.researchsquare.com/files/rs-126929/v1_stamped.pdf"}],"financialInterests":"","formattedTitle":"Identification of Coronary Calcifications in Optical Coherence Tomography Imaging Using Deep Learning","fulltext":[{"header":"Full Text","content":"\u003cp\u003eThis preprint is available for \u003ca href='/article/rs-126929/latest.pdf' target='_blank'\u003edownload as a PDF\u003c/a\u003e.\u003c/p\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":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Optical coherence tomography, Coronary calcifications, Deep Learning","lastPublishedDoi":"10.21203/rs.3.rs-126929/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-126929/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eCoronary calcifications are an obstacle for successful percutaneous treatment of coronary artery disease patients. The optimal method for delineating calcifications extent is optical coherence tomography (coronary OCT). To identify calcification on OCT and subsequently tailor the appropriate treatment, requires expertise in both image acquisition and interpretation. Image acquisition consists from system calibration, blood clearance by a contrast agent along with synchronization of the pullback process. Accurate interpretation demands careful review by the operator of a segment of 50-75mm of the coronary vessel at steps of 0.5-1mm accounting for 75-100 images in each OCT run, which is time consuming and necessitates some expertise in OCT analysis.\u003c/p\u003e\u003cp\u003eIn this paper we developed a new deep learning algorithm to assist the physician to identify and quantify coronary calcifications promptly, efficiently and accurately. Our algorithm achieves an accuracy of 0.9903 ± 0.009 over the test set at size of 1500 frames and even managed to find calcifications that weren’t recognized manually by the physician.\u0026nbsp;For the best knowledge of the authors our algorithm achieves high accuracy which was never achieved in the past.\u003c/p\u003e","manuscriptTitle":"Identification of Coronary Calcifications in Optical Coherence Tomography Imaging Using Deep Learning","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2020-12-21 21:58:53","doi":"10.21203/rs.3.rs-126929/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Major revision","date":"2021-02-25T06:28:25+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2021-02-23T02:57:50+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2021-02-22T13:20:32+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"f40e9a97-8efc-4fc9-af9d-6f0ead2baac8","date":"2021-02-14T05:07:59+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"a62bcc7e-0b60-4bf5-a2fb-91b27fb0f240","date":"2021-02-14T01:50:55+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2020-12-24T06:59:53+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2020-12-18T02:14:33+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2020-12-17T15:33:12+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2020-12-17T07:14:23+00:00","index":"","fulltext":""},{"type":"submitted","content":"Scientific Reports","date":"2020-12-11T20:03:37+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"e335b7b6-41cc-4305-afb3-009323b4fda2","owner":[],"postedDate":"December 21st, 2020","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":1575969,"name":"Nuclear Medicine \u0026 Medical Imaging"},{"id":1575970,"name":"Bioinformatics"},{"id":1575971,"name":"Cardiac \u0026 Cardiovascular Systems"},{"id":1575972,"name":"Computational Biology"}],"tags":[],"updatedAt":"2021-05-13T04:44:06+00:00","versionOfRecord":[],"versionCreatedAt":"2020-12-21 21:58:53","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-126929","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-126929","identity":"rs-126929","version":["v1"]},"buildId":"_2-kVJe1T_tPrBINL-cwx","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}