Computer-Aided Detection (CAD) Software Versus Radiologists from Multiple Countries: A Comparison of Tuberculosis Detection from Chest X-Rays

Computer-Aided Detection (CAD) Software Versus Radiologists from Multiple Countries: A Comparison of Tuberculosis Detection from Chest X-Rays | 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 Computer-Aided Detection (CAD) Software Versus Radiologists from Multiple Countries: A Comparison of Tuberculosis Detection from Chest X-Rays Zhi Zhen Qin, Martie Van der Walt, Sizulu Moyo, Farzana Ismail, and 12 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-5882564/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 02 Jul, 2025 Read the published version in Scientific Reports → Version 1 posted 11 You are reading this latest preprint version Abstract Nearly a third of TB cases go undetected annually. WHO recommends computer-aided detection (CAD) to enhance TB screening, with studies showing comparable performance to local radiologists. Using 774 chest X-rays from the South African National TB Prevalence Survey, we compared 12 CAD software with 11 radiologists from Nigeria, India, the UK, and the US, against a composite microbiological reference standard. Sensitivity, specificity and Cohen’s kappa were calculated and compared. Receiver-operating characteristic curves were developed for CAD and Euclidean distance assessed radiologists’ alignment with the best-performing software. Binomial regression tested the impact of radiologists’ characteristics on accuracy. Radiologist performance varied. On the restricted read, British radiologists had the highest sensitivity (78.7% [73.2–83.5%]) and Indian radiologists the lowest (67.1% [61.0-72.8%]). Specificity ranged from 75.8% (71.8–79.4%, Nigeria) to 84.3% (80.9–87.3%, the US). The top CAD outperformed all except Indian radiologists when matching specificity. CAD with Conformité Européenne (CE) generally matched or surpassed radiologists. British radiologists’ sensitivity was closest to the top CAD, while American radiologists were closest in specificity and overall. Experience, TB reads, and country had no significant impact on accuracy. CAD performed well against radiologists globally, underscoring its potential to enhance access to care. Health sciences/Diseases/Infectious diseases/Tuberculosis Health sciences/Health care/Medical imaging Biological sciences/Computational biology and bioinformatics/Machine learning Full Text Additional Declarations No competing interests reported. Supplementary Files CADvsRadiologistsAnnex.docx Cite Share Download PDF Status: Published Journal Publication published 02 Jul, 2025 Read the published version in Scientific Reports → Version 1 posted Editorial decision: Revision requested 15 Apr, 2025 Reviews received at journal 09 Apr, 2025 Reviewers agreed at journal 30 Mar, 2025 Reviewers agreed at journal 23 Feb, 2025 Reviews received at journal 18 Feb, 2025 Reviewers agreed at journal 10 Feb, 2025 Reviewers invited by journal 28 Jan, 2025 Editor assigned by journal 26 Jan, 2025 Editor invited by journal 25 Jan, 2025 Submission checks completed at journal 23 Jan, 2025 First submitted to journal 22 Jan, 2025 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-5882564","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":407185250,"identity":"627c8234-feaa-4064-b07c-fd3cdf6a2520","order_by":0,"name":"Zhi Zhen Qin","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA40lEQVRIiWNgGAWjYDACCcYGZhDND+EeIEGLJJhKIEoLAwNYi8EBYrXIz25u/lxQcUd28/nzBx8X/rjDwC99/AJeLYxzDjYYzzjzzHjbjWRm4xkJzxgk+3IK8GphlkhsSOZtO5y47QYzmzRPwmEGgzM8CXi1sAG1HOb9dzhxc/9hIrXwSCQ2NvM2HE7cwJAM08J+AK8WCYnEZmaeY4eNZ9xINjaekfaMR7KHB68OBvkZ6Y8/89Qclu3vP/jwcYHNHTl+HvYH+PVAAWMDAySCgFbwGJCmBQiItGUUjIJRMApGDAAAwedIm9i9TbAAAAAASUVORK5CYII=","orcid":"","institution":"Stop TB Partnership","correspondingAuthor":true,"prefix":"","firstName":"Zhi","middleName":"Zhen","lastName":"Qin","suffix":""},{"id":407185251,"identity":"591505f1-5640-4681-b971-07a1fad48532","order_by":1,"name":"Martie Van der Walt","email":"","orcid":"","institution":"South Africa Medical Research Council","correspondingAuthor":false,"prefix":"","firstName":"Martie","middleName":"Van der","lastName":"Walt","suffix":""},{"id":407185252,"identity":"6847f6dc-0f9c-42e5-9476-cec19652e218","order_by":2,"name":"Sizulu Moyo","email":"","orcid":"","institution":"Human Sciences Research Council","correspondingAuthor":false,"prefix":"","firstName":"Sizulu","middleName":"","lastName":"Moyo","suffix":""},{"id":407185253,"identity":"02b100c1-2bb0-4800-a3b0-c0f3f327414c","order_by":3,"name":"Farzana Ismail","email":"","orcid":"","institution":"National Institute for Communicable diseases","correspondingAuthor":false,"prefix":"","firstName":"Farzana","middleName":"","lastName":"Ismail","suffix":""},{"id":407185254,"identity":"b61267ef-8ed7-4819-a402-083f4f673539","order_by":4,"name":"Phaleng Maribe","email":"","orcid":"","institution":"Human Sciences Research Council","correspondingAuthor":false,"prefix":"","firstName":"Phaleng","middleName":"","lastName":"Maribe","suffix":""},{"id":407185255,"identity":"03598e50-c9b2-4a3f-8168-b66bafdfe934","order_by":5,"name":"Claudia Denkinger","email":"","orcid":"","institution":"Heidelberg University Hospital","correspondingAuthor":false,"prefix":"","firstName":"Claudia","middleName":"","lastName":"Denkinger","suffix":""},{"id":407185257,"identity":"b51da290-0d03-40c8-a172-42f4800e2efd","order_by":6,"name":"Sarah Zaidi","email":"","orcid":"","institution":"Stop TB Partnership","correspondingAuthor":false,"prefix":"","firstName":"Sarah","middleName":"","lastName":"Zaidi","suffix":""},{"id":407185258,"identity":"78a91a6a-0453-4b15-b057-cd70c22a7a84","order_by":7,"name":"Rachael Barrett","email":"","orcid":"","institution":"Stop TB Partnership","correspondingAuthor":false,"prefix":"","firstName":"Rachael","middleName":"","lastName":"Barrett","suffix":""},{"id":407185261,"identity":"b8a6f019-b1c6-4514-a16b-bb7d2d78f6f8","order_by":8,"name":"Lindiwe Mvusi","email":"","orcid":"","institution":"South African National Department of Health","correspondingAuthor":false,"prefix":"","firstName":"Lindiwe","middleName":"","lastName":"Mvusi","suffix":""},{"id":407185263,"identity":"bee1bfc1-2916-44df-b916-c86ded90dbf6","order_by":9,"name":"Yolisa Tsibolane","email":"","orcid":"","institution":"South African National Department of Health","correspondingAuthor":false,"prefix":"","firstName":"Yolisa","middleName":"","lastName":"Tsibolane","suffix":""},{"id":407185265,"identity":"dc9311a7-d1c7-4082-896e-203a2c3480ae","order_by":10,"name":"Nkateko Mkhondo","email":"","orcid":"","institution":"World Health Organization","correspondingAuthor":false,"prefix":"","firstName":"Nkateko","middleName":"","lastName":"Mkhondo","suffix":""},{"id":407185267,"identity":"8131c611-3b87-4b9c-a3e9-3993656539f2","order_by":11,"name":"Khangelani Zuma","email":"","orcid":"","institution":"Human Sciences Research Council","correspondingAuthor":false,"prefix":"","firstName":"Khangelani","middleName":"","lastName":"Zuma","suffix":""},{"id":407185268,"identity":"2c8e63c6-f156-4457-9fb7-96ebddc9e6af","order_by":12,"name":"Samuel Manda","email":"","orcid":"","institution":"South Africa Medical Research Council","correspondingAuthor":false,"prefix":"","firstName":"Samuel","middleName":"","lastName":"Manda","suffix":""},{"id":407185269,"identity":"d12a8689-8836-40af-885c-cdccf30f98e2","order_by":13,"name":"Lisa Koeppel","email":"","orcid":"","institution":"Heidelberg University Hospital","correspondingAuthor":false,"prefix":"","firstName":"Lisa","middleName":"","lastName":"Koeppel","suffix":""},{"id":407185270,"identity":"24f2ee03-ea4a-400e-ae48-50ab6a95d593","order_by":14,"name":"Thuli Mthiyane","email":"","orcid":"","institution":"South Africa Medical Research Council","correspondingAuthor":false,"prefix":"","firstName":"Thuli","middleName":"","lastName":"Mthiyane","suffix":""},{"id":407185271,"identity":"7b2e1123-2c05-4daf-9396-c81c03f06c1f","order_by":15,"name":"Jacob Creswell","email":"","orcid":"","institution":"Stop TB Partnership","correspondingAuthor":false,"prefix":"","firstName":"Jacob","middleName":"","lastName":"Creswell","suffix":""}],"badges":[],"createdAt":"2025-01-22 16:38:24","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-5882564/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-5882564/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1038/s41598-025-06164-w","type":"published","date":"2025-07-02T15:57:26+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":86178977,"identity":"d4f49492-ed05-4719-9237-5fc7a57f89c8","added_by":"auto","created_at":"2025-07-07 16:13:53","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":646994,"visible":true,"origin":"","legend":"","description":"","filename":"CleanCADSoftwareVersusRadiologistsfromMultipleCountriesAComparisonofTuberculosisDetectionfromChestXRaysCopy.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5882564/v1_covered_3f3613fe-cf66-4b3a-a803-30dd8b91a382.pdf"},{"id":74959799,"identity":"cf8cd482-83f5-46db-becd-bbf5493030a1","added_by":"auto","created_at":"2025-01-28 18:40:14","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":91183,"visible":true,"origin":"","legend":"","description":"","filename":"CADvsRadiologistsAnnex.docx","url":"https://assets-eu.researchsquare.com/files/rs-5882564/v1/ae0e8279c37448b2513071eb.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Computer-Aided Detection (CAD) Software Versus Radiologists from Multiple Countries: A Comparison of Tuberculosis Detection from Chest X-Rays","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":false,"isPdf":true,"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":"","lastPublishedDoi":"10.21203/rs.3.rs-5882564/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-5882564/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eNearly a third of TB cases go undetected annually. WHO recommends computer-aided detection (CAD) to enhance TB screening, with studies showing comparable performance to local radiologists. Using 774 chest X-rays from the South African National TB Prevalence Survey, we compared 12 CAD software with 11 radiologists from Nigeria, India, the UK, and the US, against a composite microbiological reference standard. Sensitivity, specificity and Cohen\u0026rsquo;s kappa were calculated and compared. Receiver-operating characteristic curves were developed for CAD and Euclidean distance assessed radiologists\u0026rsquo; alignment with the best-performing software. Binomial regression tested the impact of radiologists\u0026rsquo; characteristics on accuracy. Radiologist performance varied. On the restricted read, British radiologists had the highest sensitivity (78.7% [73.2\u0026ndash;83.5%]) and Indian radiologists the lowest (67.1% [61.0-72.8%]). Specificity ranged from 75.8% (71.8\u0026ndash;79.4%, Nigeria) to 84.3% (80.9\u0026ndash;87.3%, the US). The top CAD outperformed all except Indian radiologists when matching specificity. CAD with \u003cem\u003eConformit\u0026eacute; Europ\u0026eacute;enne\u003c/em\u003e (CE) generally matched or surpassed radiologists. British radiologists\u0026rsquo; sensitivity was closest to the top CAD, while American radiologists were closest in specificity and overall. Experience, TB reads, and country had no significant impact on accuracy. CAD performed well against radiologists globally, underscoring its potential to enhance access to care.\u003c/p\u003e","manuscriptTitle":"Computer-Aided Detection (CAD) Software Versus Radiologists from Multiple Countries: A Comparison of Tuberculosis Detection from Chest X-Rays","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-01-28 18:32:09","doi":"10.21203/rs.3.rs-5882564/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-04-15T07:53:46+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-04-09T11:00:13+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"311444673106335029255675562778812221932","date":"2025-03-30T07:06:16+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"224965823627776707118678894381150814018","date":"2025-02-23T07:48:35+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-02-18T16:03:14+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"330743516832378685993451389834840974580","date":"2025-02-11T03:49:37+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-01-28T09:17:21+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-01-26T06:27:49+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2025-01-25T13:51:03+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-01-23T13:20:55+00:00","index":"","fulltext":""},{"type":"submitted","content":"Scientific Reports","date":"2025-01-22T16:36:18+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":"34d023ca-258c-45e6-821a-ac0fe75ce4cb","owner":[],"postedDate":"January 28th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[{"id":43422172,"name":"Health sciences/Diseases/Infectious diseases/Tuberculosis"},{"id":43422173,"name":"Health sciences/Health care/Medical imaging"},{"id":43422174,"name":"Biological sciences/Computational biology and bioinformatics/Machine learning"}],"tags":[],"updatedAt":"2025-07-07T16:01:45+00:00","versionOfRecord":{"articleIdentity":"rs-5882564","link":"https://doi.org/10.1038/s41598-025-06164-w","journal":{"identity":"scientific-reports","isVorOnly":false,"title":"Scientific Reports"},"publishedOn":"2025-07-02 15:57:26","publishedOnDateReadable":"July 2nd, 2025"},"versionCreatedAt":"2025-01-28 18:32:09","video":"","vorDoi":"10.1038/s41598-025-06164-w","vorDoiUrl":"https://doi.org/10.1038/s41598-025-06164-w","workflowStages":[]},"version":"v1","identity":"rs-5882564","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-5882564","identity":"rs-5882564","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}