ParaPET: non-invasive deep learning method for direct parametric PET reconstruction using histoimages

ParaPET: non-invasive deep learning method for direct parametric PET reconstruction using histoimages | 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 ParaPET: non-invasive deep learning method for direct parametric PET reconstruction using histoimages Rajat Vashistha, Hamed Moradi, Amanda Hammond, Kieran O’Brien, and 5 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-3311784/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 30 Jan, 2024 Read the published version in EJNMMI Research → Version 1 posted 3 You are reading this latest preprint version Abstract Background The indirect method for generating parametric images in Positron Emission Tomography (PET) involves the acquisition and reconstruction of dynamic images and temporal modelling of tissue activity given a measured arterial input function. This approach is not robust, as noise in each dynamic image leads to a degradation in parameter estimation. Direct methods incorporate into the image reconstruction step both the kinetic and noise models, leading to improved parametric images. These methods require extensive computational time and large computing resources. Machine learning methods have demonstrated significant potential in overcoming these challenges. but they are limited by the requirement of a paired training dataset. A further challenge within the existing framework is the use of state-of-the-art arterial input function estimation via temporal arterial blood sampling, which is an invasive procedure, or an additional Magnetic Resonance Imaging (MRI) scan for selecting a region where arterial blood signal can be measured from the PET image. We propose a novel machine learning approach for reconstructing high-quality parametric images from histoimages produced from time-of-flight PET data without requiring invasive arterial sampling, MRI scan or paired training data. Result The proposed is tested on a simulated phantom and five oncological subjects undergoing an 18F-FDG-PET scan of the brain using Siemens Biograph Vision Quadra. Kinetic parameters set in the brain phantom correlated strongly with the estimated parameters ( K 1 , k 2 and k 3 , Pearson correlation coefficient of 0.91, 0.92 and 0.93) and a mean squared error of less than 0.0004. In addition, our method significantly outperforms (p < 0.05, paired t-test) the conventional non-linear least squares method in terms of contrast-to-noise ratio. At last, the proposed method was found to be 37% faster than the conventional method. Conclusion We proposed a direct non-invasive DL-based reconstruction method producing parametric images of higher quality. The use of histoimages holds promising potential for enhancing the estimation of parametric images, an area that has not been extensively explored thus far. The proposed method can be applied to subject-specific dynamic PET data alone. Positron Emission Tomography direct parametric image reconstruction deep learning histoimages Full Text Cite Share Download PDF Status: Published Journal Publication published 30 Jan, 2024 Read the published version in EJNMMI Research → Version 1 posted Editor invited by journal 08 Sep, 2023 Editor assigned by journal 31 Aug, 2023 First submitted to journal 30 Aug, 2023 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-3311784","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":251000515,"identity":"ac3e7dbb-57e7-4fd6-9fd5-7588afb08dc8","order_by":0,"name":"Rajat Vashistha","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA7klEQVRIiWNgGAWjYLACxgYJBgYJBsYHYN4BYrXwSDAwG5CihQGkhU2CKC3mDTyGH3/usLC3l25+Vs1TxiDHdyOB8TMPHi0yB3iMpXnPSCT2yBwzu81zjsFY8kYCszQ+LUDnJEgztkkk8EgkmN3mbWNI3HAjgYGQluSfP9sk7Hkk0r8VA7XUA7Uw/8avhfmYBG+bBGOPRI4ZM1BLgsGNBDb8tjAzH7MGaknsuZFTLDnnnIThzDMP2yzn4NPC3th882dbnT37jPSNH96U2cjzHU8+fOMNHi0MzCg8cNSAool4wEaK4lEwCkbBKBgpAADlB0ISgkIKZgAAAABJRU5ErkJggg==","orcid":"https://orcid.org/0000-0001-6800-7661","institution":"University of Queensland Centre for Advanced Imaging","correspondingAuthor":true,"submittingAuthor":false,"prefix":"","firstName":"Rajat","middleName":"","lastName":"Vashistha","suffix":""},{"id":251000516,"identity":"7f7cc10c-2a25-44b0-9b9d-ae806ac23eff","order_by":1,"name":"Hamed Moradi","email":"","orcid":"","institution":"The University of Queensland Centre for Advanced Imaging","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Hamed","middleName":"","lastName":"Moradi","suffix":""},{"id":251000517,"identity":"fe8d7dd9-4340-4453-a1ee-4d08f91e57a2","order_by":2,"name":"Amanda Hammond","email":"","orcid":"","institution":": Siemens Healthcare Private Limited","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Amanda","middleName":"","lastName":"Hammond","suffix":""},{"id":251000518,"identity":"bd65d3f8-fdd1-4451-bf50-4559ca794723","order_by":3,"name":"Kieran O’Brien","email":"","orcid":"","institution":"Siemens Healthcare Private Limited","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Kieran","middleName":"","lastName":"O’Brien","suffix":""},{"id":251000519,"identity":"f4eb18de-842d-46d9-98c2-17b69f8529a5","order_by":4,"name":"Axel Rominger","email":"","orcid":"","institution":"University of Bern: Universitat Bern","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Axel","middleName":"","lastName":"Rominger","suffix":""},{"id":251000520,"identity":"e74273bf-3915-48f6-89d0-cf33f903fdd9","order_by":5,"name":"Hasan Sari","email":"","orcid":"","institution":"University of Bern: Universitat Bern","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Hasan","middleName":"","lastName":"Sari","suffix":""},{"id":251000521,"identity":"769f9f60-8534-4fc4-8663-9ca5bc4af5b5","order_by":6,"name":"Kuangyu Shi","email":"","orcid":"","institution":"University of Bern: Universitat Bern","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Kuangyu","middleName":"","lastName":"Shi","suffix":""},{"id":251000522,"identity":"52ed6464-40f9-446e-b540-bd69131aa8f1","order_by":7,"name":"Viktor Vegh","email":"","orcid":"https://orcid.org/0000-0002-4310-1421","institution":"The University of Queensland Centre for Advanced Imaging","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Viktor","middleName":"","lastName":"Vegh","suffix":""},{"id":251000523,"identity":"e03a60a2-68bd-4966-ba90-5eb98dc7b65f","order_by":8,"name":"David Reutens","email":"","orcid":"","institution":"The University of Queensland Centre for Advanced Imaging","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"David","middleName":"","lastName":"Reutens","suffix":""}],"badges":[],"createdAt":"2023-08-31 01:46:27","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-3311784/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-3311784/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1186/s13550-024-01072-y","type":"published","date":"2024-01-30T15:00:36+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":50674440,"identity":"9e054d49-6015-4e5b-8439-1425e9221faa","added_by":"auto","created_at":"2024-02-05 15:10:58","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":450529,"visible":true,"origin":"","legend":"","description":"","filename":"ParaPETEJNMMIRES.pdf","url":"https://assets-eu.researchsquare.com/files/rs-3311784/v1_covered_7746a730-3c4d-4e0e-b94c-016fa87761d5.pdf"}],"financialInterests":"","formattedTitle":"ParaPET: non-invasive deep learning method for direct parametric PET reconstruction using histoimages","fulltext":[],"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":true,"isPdf":true,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"ejnmmi-research","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"ejre","sideBox":"Learn more about [EJNMMI Research](http://ejnmmires.springeropen.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/ejre/default.aspx","title":"EJNMMI Research","twitterHandle":"@officialEANM","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"BMC/SO AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Positron Emission Tomography, direct parametric image reconstruction, deep learning, histoimages","lastPublishedDoi":"10.21203/rs.3.rs-3311784/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-3311784/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eBackground\u003c/h2\u003e \u003cp\u003eThe indirect method for generating parametric images in Positron Emission Tomography (PET) involves the acquisition and reconstruction of dynamic images and temporal modelling of tissue activity given a measured arterial input function. This approach is not robust, as noise in each dynamic image leads to a degradation in parameter estimation. Direct methods incorporate into the image reconstruction step both the kinetic and noise models, leading to improved parametric images. These methods require extensive computational time and large computing resources. Machine learning methods have demonstrated significant potential in overcoming these challenges. but they are limited by the requirement of a paired training dataset. A further challenge within the existing framework is the use of state-of-the-art arterial input function estimation via temporal arterial blood sampling, which is an invasive procedure, or an additional Magnetic Resonance Imaging (MRI) scan for selecting a region where arterial blood signal can be measured from the PET image. We propose a novel machine learning approach for reconstructing high-quality parametric images from histoimages produced from time-of-flight PET data without requiring invasive arterial sampling, MRI scan or paired training data.\u003c/p\u003e\u003ch2\u003eResult\u003c/h2\u003e \u003cp\u003eThe proposed is tested on a simulated phantom and five oncological subjects undergoing an 18F-FDG-PET scan of the brain using Siemens Biograph Vision Quadra. Kinetic parameters set in the brain phantom correlated strongly with the estimated parameters (\u003cem\u003eK\u003c/em\u003e\u003csub\u003e\u003cem\u003e1\u003c/em\u003e\u003c/sub\u003e, \u003cem\u003ek\u003c/em\u003e\u003csub\u003e\u003cem\u003e2\u003c/em\u003e\u003c/sub\u003e and \u003cem\u003ek\u003c/em\u003e\u003csub\u003e\u003cem\u003e3\u003c/em\u003e\u003c/sub\u003e, Pearson correlation coefficient of 0.91, 0.92 and 0.93) and a mean squared error of less than 0.0004. In addition, our method significantly outperforms (p\u0026thinsp;\u0026lt;\u0026thinsp;0.05, paired t-test) the conventional non-linear least squares method in terms of contrast-to-noise ratio. At last, the proposed method was found to be 37% faster than the conventional method.\u003c/p\u003e\u003ch2\u003eConclusion\u003c/h2\u003e \u003cp\u003eWe proposed a direct non-invasive DL-based reconstruction method producing parametric images of higher quality. The use of histoimages holds promising potential for enhancing the estimation of parametric images, an area that has not been extensively explored thus far. The proposed method can be applied to subject-specific dynamic PET data alone.\u003c/p\u003e","manuscriptTitle":"ParaPET: non-invasive deep learning method for direct parametric PET reconstruction using histoimages","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2023-11-21 11:10:26","doi":"10.21203/rs.3.rs-3311784/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"editorInvited","content":"EJNMMI Research","date":"2023-09-08T08:56:55+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2023-09-01T00:01:01+00:00","index":"","fulltext":""},{"type":"submitted","content":"EJNMMI Research","date":"2023-08-31T02:22:49+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"ejnmmi-research","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"ejre","sideBox":"Learn more about [EJNMMI Research](http://ejnmmires.springeropen.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/ejre/default.aspx","title":"EJNMMI Research","twitterHandle":"@officialEANM","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"BMC/SO AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"c5c377d3-8f3b-43c2-9fde-35e204e64611","owner":[],"postedDate":"November 21st, 2023","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2024-02-05T15:08:38+00:00","versionOfRecord":{"articleIdentity":"rs-3311784","link":"https://doi.org/10.1186/s13550-024-01072-y","journal":{"identity":"ejnmmi-research","isVorOnly":false,"title":"EJNMMI Research"},"publishedOn":"2024-01-30 15:00:36","publishedOnDateReadable":"January 30th, 2024"},"versionCreatedAt":"2023-11-21 11:10:26","video":"","vorDoi":"10.1186/s13550-024-01072-y","vorDoiUrl":"https://doi.org/10.1186/s13550-024-01072-y","workflowStages":[]},"version":"v1","identity":"rs-3311784","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-3311784","identity":"rs-3311784","version":["v1"]},"buildId":"FbvkV6FR0MCFSLy54lSbu","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}