Differential Diagnosis of Pulmonary Arterial Hypertension via Deep Multimodal Modeling of Noninvasive Measurements

Differential Diagnosis of Pulmonary Arterial Hypertension via Deep Multimodal Modeling of Noninvasive Measurements | 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 Differential Diagnosis of Pulmonary Arterial Hypertension via Deep Multimodal Modeling of Noninvasive Measurements Siqiang Zheng, Enkuo Zheng, Liang Guo, Lei Xie, Bin Zhao, Qi Hong, and 2 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-5652320/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 Background Pulmonary hypertension (PH) is a rare yet progressive disorder characterized by elevated blood pressure in the arteries of the lungs, often presenting without an evident cause. The distinction between pulmonary arterial hypertension (PAH) and PH due to left heart disease (PH-LHD) is crucial due to differences in treatment approach and prognosis. A correct diagnosis is essential to ensuring that patients receive the appropriate treatment. However, the current diagnosis of PAH and PH-LHD faces clinical limitations due to its reliance on invasive hemodynamic measurements or the limits of data modeling. Methods We conducted a retrospective study involving 905 PH patients from Shanghai Pulmonary Hospital, incorporating CT images and clinical textual data. A heterogeneous graph neural network was constructed on multimodal data. We developed and validated a heterogeneous network to predict the probability of PAH and PH-LHD. The new method exhibits strong discrimination between PAH and PH-LHD. Findings In the multiple cross-validations, it achieved a 0.943 AUC when using a combination of image and textual data, a 0.869 AUC for only textual variables, and a 0.715 AUC for only images. It also evaluated the type-dependent performances, yielding the best AUC (0.896) when using a subgroup of variables. The best performance (AUC=0.809) can be achieved by using only ECG variables in our model. Interpretation Our study provides a quantitative rationale for modulating multimodal data to enhance differential diagnosis of PAH and PH-LHD with only noninvasive measurements. Deep Learning Pulmonary Arterial Hypertension Pulmonary Hypertension due to Left Heart Disease Graph Neural Network Full Text Additional Declarations No competing interests reported. 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-5652320","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":391523685,"identity":"f4c49cc8-eb09-4404-a607-fcea40ebc3ce","order_by":0,"name":"Siqiang Zheng","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA4ElEQVRIiWNgGAWjYDACZgaGAyDagL0BRFqQooUHRBlIkGCbgUQCiCJCi3w788YDP3fU2ptLPr+64UeBBAN/e3cCfsMPsxUc7D1znNlydk7ZzR6gwyTOnN2AXwszj8EB3rZjbAa3c9Ju8AC1GEjk4tci38xjcPBv2zEeg5tn0m7+IUYLw2Eeg8O8bTUSBjfYj90myhaQXw7Lth0wMDiTw3ZbxkCCh6Bf5PsPb/74tq3O3uD48Wc33/yxkeNv7yXgMKBFIOcBMQ+IwcBDSDlMSx0Qsz8gRvUoGAWjYBSMQAAAqN9JfhMzMlkAAAAASUVORK5CYII=","orcid":"","institution":"Shanghai Pulmonary Hospital","correspondingAuthor":true,"prefix":"","firstName":"Siqiang","middleName":"","lastName":"Zheng","suffix":""},{"id":391523686,"identity":"83dc6191-6947-4ae6-aea4-6def878264b2","order_by":1,"name":"Enkuo Zheng","email":"","orcid":"","institution":"Second Hospital of Ningbo","correspondingAuthor":false,"prefix":"","firstName":"Enkuo","middleName":"","lastName":"Zheng","suffix":""},{"id":391523687,"identity":"e059698a-c998-4856-bfe5-8560f2d3f65a","order_by":2,"name":"Liang Guo","email":"","orcid":"","institution":"Shanghai Pulmonary Hospital","correspondingAuthor":false,"prefix":"","firstName":"Liang","middleName":"","lastName":"Guo","suffix":""},{"id":391523688,"identity":"ac04e94f-b9bb-4389-8b09-a2a00ac893ab","order_by":3,"name":"Lei Xie","email":"","orcid":"","institution":"Second Affiliated Hospital of Nanchang University","correspondingAuthor":false,"prefix":"","firstName":"Lei","middleName":"","lastName":"Xie","suffix":""},{"id":391523689,"identity":"6af3b5ee-6bed-4470-859c-7c0e79e27902","order_by":4,"name":"Bin Zhao","email":"","orcid":"","institution":"Shandong Public Health Clinical Center","correspondingAuthor":false,"prefix":"","firstName":"Bin","middleName":"","lastName":"Zhao","suffix":""},{"id":391523690,"identity":"8aaeb9d6-0e7d-4a92-b3da-50ffbac4df5c","order_by":5,"name":"Qi Hong","email":"","orcid":"","institution":"Shenyang Tenth People’s Hospital, Shenyang Chest Hospital","correspondingAuthor":false,"prefix":"","firstName":"Qi","middleName":"","lastName":"Hong","suffix":""},{"id":391523691,"identity":"c07d6938-f0be-4d2a-8084-f014917c8cdc","order_by":6,"name":"Xuefei Hu","email":"","orcid":"","institution":"Shanghai Pulmonary Hospital","correspondingAuthor":false,"prefix":"","firstName":"Xuefei","middleName":"","lastName":"Hu","suffix":""},{"id":391523692,"identity":"18fa853f-d212-4591-a0fb-6f02608a1ca3","order_by":7,"name":"Bo Tao","email":"","orcid":"","institution":"Shanghai Pulmonary Hospital","correspondingAuthor":false,"prefix":"","firstName":"Bo","middleName":"","lastName":"Tao","suffix":""}],"badges":[],"createdAt":"2024-12-16 09:23:23","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-5652320/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-5652320/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":72530252,"identity":"34c220ac-9d92-44b8-8deb-af318c8a642c","added_by":"auto","created_at":"2024-12-28 19:16:31","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":3379580,"visible":true,"origin":"","legend":"","description":"","filename":"PAH.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5652320/v1_covered_d4200c95-f110-49a1-8b95-2233a9621d8a.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Differential Diagnosis of Pulmonary Arterial Hypertension via Deep Multimodal Modeling of Noninvasive Measurements","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":"Deep Learning, Pulmonary Arterial Hypertension, Pulmonary Hypertension due to Left Heart Disease, Graph Neural Network","lastPublishedDoi":"10.21203/rs.3.rs-5652320/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-5652320/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cstrong\u003eBackground\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003ePulmonary hypertension (PH) is a rare yet progressive disorder characterized by elevated blood pressure in the arteries of the lungs, often presenting without an evident cause. The distinction between pulmonary arterial hypertension (PAH) and PH due to left heart disease (PH-LHD) is crucial due to differences in treatment approach and prognosis. A correct diagnosis is essential to ensuring that patients receive the appropriate treatment. However, the current diagnosis of PAH and PH-LHD faces clinical limitations due to its reliance on invasive hemodynamic measurements or the limits of data modeling.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMethods\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe conducted a retrospective study involving 905 PH patients from Shanghai Pulmonary Hospital, incorporating CT images and clinical textual data. A heterogeneous graph neural network was constructed on multimodal data. We developed and validated a heterogeneous network to predict the probability of PAH and PH-LHD. The new method exhibits strong discrimination between PAH and PH-LHD.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFindings\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eIn the multiple cross-validations, it achieved a 0.943 AUC when using a combination of image and textual data, a 0.869 AUC for only textual variables, and a 0.715 AUC for only images. It also evaluated the type-dependent performances, yielding the best AUC (0.896) when using a subgroup of variables. The best performance (AUC=0.809) can be achieved by using only ECG variables in our model.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eInterpretation\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eOur study provides a quantitative rationale for modulating multimodal data to enhance differential diagnosis of PAH and PH-LHD with only noninvasive measurements.\u003c/p\u003e","manuscriptTitle":"Differential Diagnosis of Pulmonary Arterial Hypertension via Deep Multimodal Modeling of Noninvasive Measurements","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-12-19 11:50:42","doi":"10.21203/rs.3.rs-5652320/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":"07389f03-df48-431b-b807-8efc2828066b","owner":[],"postedDate":"December 19th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2024-12-28T19:08:10+00:00","versionOfRecord":[],"versionCreatedAt":"2024-12-19 11:50:42","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-5652320","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-5652320","identity":"rs-5652320","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}