Optimization of a Q.Clear PET Image Reconstruction Based on Bayesian Optimization in Oncology and Neurology Applications

Optimization of a Q.Clear PET Image Reconstruction Based on Bayesian Optimization in Oncology and Neurology Applications | 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 Optimization of a Q.Clear PET Image Reconstruction Based on Bayesian Optimization in Oncology and Neurology Applications Qariemah Azahar, Hazlin Hashim, Khadijah Abdul Hamid, Lyu Xin, and 1 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6912272/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 Objective : The Bayesian Penalized Likelihood (BPL) reconstruction algorithm in PET/CT imaging has shown potential for enhancing the image quality and quantification. This study aims to determine the optimal Beta (β) value in Q.Clear reconstruction using Bayesian Optimization (BO) through phantoms and clinical evaluation. Methods : Both phantoms (NEMA IQ and Hoffman Brain) and clinical images were evaluated. The images of both phantoms were reconstructed using different β values from Q50 to Q1000 through different acquisition times. The RC% and CNR was calculated on both phantoms for quantification evaluation. For the clinical studies, one neurology and one oncology patient data were selected and reconstructed using OSEM and Q.Clear different b values from Q100 to Q1000, with full-time and reduce acquisition time. Visual assessment was performed independently by two nuclear medicine physicians, and interrater agreement was analyzed using Krippendorff’s alpha. Results : For the NEMA IQ phantom, increasing β values and acquisition times resulted in decreased RC% and increased CNR. For the Hoffman Brain phantom, the cortical regions showed stable RC% values from mid to high β values, whereas the deeper brain structures were more sensitive to the changes of β, showing a reduction in RC%. Meanwhile, CNR improved with increasing b values and longer acquisition times. Optimization for RC% and CNR indicates that smaller spheres performed optimally at lower β and longer acquisition times, while larger spheres showed good tolerance at β between Q600 to Q800. For the Hoffman phantom, cortical regions performed best at β values between Q400 to Q600, while subcortical regions performed best at lower bvalues such as Q100 to Q300. Based on the visual assessment of clinical data, the highest scores were observed for Q.Clear ranging from Q300 to Q1000 in oncology, and Q500 to Q900 in neurology. Conclusion : The results demonstrated that higher β value improved RC% stability and CNR in phantom studies, with BO analysis suggesting Q500 to Q900 as the optimal range. This aligns with clinical evaluations, where images reconstructed within this β range are consistently perceived with a high total score and strong interrater agreement, indicating its higher reliability for PET/CT image reconstruction. Quantitation PET/CT Bayesian Penalized Likelihood (BPL) Image reconstruction Bayesian Optimization (BO) Q.Clear Full Text 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-6912272","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":482829453,"identity":"2ab4d8ea-e41a-47a6-8ea4-7e697450a186","order_by":0,"name":"Qariemah Azahar","email":"data:image/png;base64,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","orcid":"https://orcid.org/0009-0006-6926-7602","institution":"Advanced Medical and Dental Institute, Universiti Sains Malaysia","correspondingAuthor":true,"prefix":"","firstName":"Qariemah","middleName":"","lastName":"Azahar","suffix":""},{"id":482829454,"identity":"1a52a82a-df7e-42ca-948d-c2940215cbda","order_by":1,"name":"Hazlin Hashim","email":"","orcid":"","institution":"Advanced Medical and Dental Institute, Universiti Sains Malaysia","correspondingAuthor":false,"prefix":"","firstName":"Hazlin","middleName":"","lastName":"Hashim","suffix":""},{"id":482829455,"identity":"c3bc19b0-0d2e-45b1-80b1-a50354cd0c19","order_by":2,"name":"Khadijah Abdul Hamid","email":"","orcid":"","institution":"Advanced Medical and Dental Institute, Universiti Sains Malaysia","correspondingAuthor":false,"prefix":"","firstName":"Khadijah","middleName":"Abdul","lastName":"Hamid","suffix":""},{"id":482829456,"identity":"1e9e9477-aecd-4c8e-a138-9d870d1ffd95","order_by":3,"name":"Lyu Xin","email":"","orcid":"","institution":"School of Health Sciences, Universiti Sains Malaysia","correspondingAuthor":false,"prefix":"","firstName":"Lyu","middleName":"","lastName":"Xin","suffix":""},{"id":482829457,"identity":"237ba891-5b57-423b-98f0-85ab7e31a440","order_by":4,"name":"Syahir Mansor","email":"","orcid":"https://orcid.org/0000-0001-9315-652X","institution":"Advanced Medical and Dental Institute, Universiti Sains Malaysia","correspondingAuthor":false,"prefix":"","firstName":"Syahir","middleName":"","lastName":"Mansor","suffix":""}],"badges":[],"createdAt":"2025-06-17 08:51:35","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6912272/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6912272/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":86379043,"identity":"e66a87e1-45da-40bf-937d-bc04f8452f56","added_by":"auto","created_at":"2025-07-10 03:40:09","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2975687,"visible":true,"origin":"","legend":"","description":"","filename":"OptimizationofPETImageReconstructionQARIEMAHAZAHAR.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6912272/v1_covered_7fc5e087-d49e-40b1-b982-7b6341ee578d.pdf"}],"financialInterests":"","formattedTitle":"Optimization of a Q.Clear PET Image Reconstruction Based on Bayesian Optimization in Oncology and Neurology Applications","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"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":"Quantitation, PET/CT, Bayesian Penalized Likelihood (BPL), Image reconstruction, Bayesian Optimization (BO), Q.Clear","lastPublishedDoi":"10.21203/rs.3.rs-6912272/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6912272/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cstrong\u003eObjective\u003c/strong\u003e:\u0026nbsp;The\u0026nbsp;Bayesian Penalized Likelihood (BPL) reconstruction algorithm in PET/CT imaging has shown potential for enhancing the image quality and quantification. This study aims to determine the optimal Beta (β) value in Q.Clear reconstruction using Bayesian Optimization (BO) through phantoms and clinical evaluation.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMethods\u003c/strong\u003e:\u0026nbsp;Both phantoms (NEMA IQ and Hoffman Brain) and clinical images were evaluated. The images of both phantoms were reconstructed using different\u0026nbsp;β\u0026nbsp;values from Q50 to Q1000 through different acquisition times. The RC% and CNR was calculated on both phantoms for quantification evaluation. For the clinical studies, one neurology and one oncology patient data were selected and reconstructed using OSEM and Q.Clear different\u0026nbsp;b\u0026nbsp;values from Q100 to Q1000, with full-time and reduce acquisition time. Visual assessment was performed independently by two nuclear medicine physicians, and interrater agreement was analyzed using Krippendorff’s alpha.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eResults\u003c/strong\u003e:\u0026nbsp;For the NEMA IQ phantom, increasing\u0026nbsp;β\u0026nbsp;values and acquisition times resulted in decreased RC% and increased CNR. For the Hoffman Brain phantom, the cortical regions showed stable RC% values from mid to high\u0026nbsp;β\u0026nbsp;values, whereas the deeper brain structures were more sensitive to the changes of\u0026nbsp;β, showing a reduction in RC%. Meanwhile, CNR improved with increasing\u0026nbsp;b\u0026nbsp;values and longer acquisition times. Optimization for RC% and CNR indicates that smaller spheres performed optimally at lower\u0026nbsp;β\u0026nbsp;and longer acquisition times, while larger spheres showed good tolerance at\u0026nbsp;β\u0026nbsp;between Q600 to Q800. For the Hoffman phantom, cortical regions performed best at\u0026nbsp;β\u0026nbsp;values between Q400 to Q600, while subcortical regions performed best at lower\u0026nbsp;bvalues such as Q100 to Q300.\u0026nbsp;Based on the visual assessment of clinical data, the highest scores were observed for Q.Clear ranging from Q300 to Q1000 in oncology, and Q500 to Q900 in neurology.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConclusion\u003c/strong\u003e:\u0026nbsp;The results demonstrated that higher\u0026nbsp;β\u0026nbsp;value improved RC% stability and CNR in phantom studies, with BO analysis suggesting Q500 to Q900 as the optimal range. This aligns with clinical evaluations, where images reconstructed within this\u0026nbsp;β\u0026nbsp;range are consistently perceived with a high total score and strong interrater agreement, indicating its higher reliability for PET/CT image reconstruction.\u003c/p\u003e","manuscriptTitle":"Optimization of a Q.Clear PET Image Reconstruction Based on Bayesian Optimization in Oncology and Neurology Applications","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-07-10 03:24:01","doi":"10.21203/rs.3.rs-6912272/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":"2bddd145-f285-45a0-97df-a321e3d9c062","owner":[],"postedDate":"July 10th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2025-07-10T03:24:01+00:00","versionOfRecord":[],"versionCreatedAt":"2025-07-10 03:24:01","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-6912272","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6912272","identity":"rs-6912272","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}