Score-Enhanced Eigen-CAM (SE-CAM) Gradient Free Visual Explanation Method

Score-Enhanced Eigen-CAM (SE-CAM) Gradient Free Visual Explanation Method | 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 Score-Enhanced Eigen-CAM (SE-CAM) Gradient Free Visual Explanation Method Nandani Sharma, Sandeep Chaurasia This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9434509/v1 This work is licensed under a CC BY 4.0 License Status: Under Revision Version 1 posted 10 You are reading this latest preprint version Abstract The contribution of artificial intelligence to trust-sensitive domains, such as the military and medical domains, is challenging. The main reason that artificial intelligence is challenging to accept in trust-sensitive domains is the black-box nature of deep-learning algorithms. In order to open up the Black-Box nature of deep learning algorithms, explainable artificial intelligence (XAI) comes into the picture. The XAI branch of artificial intelligence's main contribution is to provide a layman's understanding explanation to develop trust in the decisions taken by deep learning models. In this article, we propose an XAI technique using the fusion of Eigen-CAM and Score-CAM. Eigen-CAM and Score-CAM are gradient-free methods that do not suffer from gradient saturation. Therefore, we propose the Score-Enhanced Eigen-CAM (SE-CAM), the fusion of Score-CAM and Eigen-CAM. In addition to SE-CAM, the research work also trained the VGG16 architecture for multi-class classification of chest X-rays. In addition, we examined SE-CAM on the pretrained DenseNet121 using the ChestXpert dataset. The VGG16 model achieves 94.58% accuracy for the classification of COVID-19, pneumonia, lung opacity, and normal chest X-rays. To evaluate the SE-CAM performance, the % average drop and % increase in confidence were used. For the VGG16 and COVID-19 datasets, the % average drop was 18.41 and the % increase in confidence was 45.03. The SE-CAM model outperformed the base XAI models by 5.73 % on the % average drop and % increase in confidence by 7.7 %. XAI Medical image X-ray Ensemble approach local explanation Model-specific Score-CAM Eigen-CAM Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Revision Version 1 posted Editorial decision: Revision requested 22 May, 2026 Reviews received at journal 21 May, 2026 Reviewers agreed at journal 21 May, 2026 Reviews received at journal 16 May, 2026 Reviewers agreed at journal 07 May, 2026 Reviewers invited by journal 07 May, 2026 Editor invited by journal 28 Apr, 2026 Editor assigned by journal 20 Apr, 2026 Submission checks completed at journal 20 Apr, 2026 First submitted to journal 16 Apr, 2026 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-9434509","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":641327264,"identity":"e3576bc5-ed17-4550-a41f-453e280e408a","order_by":0,"name":"Nandani Sharma","email":"","orcid":"","institution":"Manipal University Jaipur","correspondingAuthor":false,"prefix":"","firstName":"Nandani","middleName":"","lastName":"Sharma","suffix":""},{"id":641327265,"identity":"5709ca72-b743-497a-8996-03b4b9cc1b8c","order_by":1,"name":"Sandeep Chaurasia","email":"data:image/png;base64,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","orcid":"","institution":"Manipal University Jaipur","correspondingAuthor":true,"prefix":"","firstName":"Sandeep","middleName":"","lastName":"Chaurasia","suffix":""}],"badges":[],"createdAt":"2026-04-16 07:23:15","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-9434509/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9434509/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":109759404,"identity":"d565e8b7-bd91-4d3e-b760-096de6ea76ae","added_by":"auto","created_at":"2026-05-22 07:26:56","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":532048,"visible":true,"origin":"","legend":"","description":"","filename":"SECAMSpringerNature.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9434509/v1_covered_bb09c8e2-0a01-4894-bc2b-4f71c91505c4.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Score-Enhanced Eigen-CAM (SE-CAM) Gradient Free Visual Explanation Method","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"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":"discover-applied-sciences","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"Learn more about [Discover Applied Sciences](https://link.springer.com/journal/42452)","snPcode":"42452","submissionUrl":"https://submission.springernature.com/new-submission/42452/3","title":"Discover Applied Sciences","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Discover Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"XAI, Medical image, X-ray, Ensemble approach, local explanation, Model-specific, Score-CAM, Eigen-CAM","lastPublishedDoi":"10.21203/rs.3.rs-9434509/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9434509/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eThe contribution of artificial intelligence to trust-sensitive domains, such as the military and medical domains, is challenging. The main reason that artificial intelligence is challenging to accept in trust-sensitive domains is the black-box nature of deep-learning algorithms. In order to open up the Black-Box nature of deep learning algorithms, explainable artificial intelligence (XAI) comes into the picture. The XAI branch of artificial intelligence's main contribution is to provide a layman's understanding explanation to develop trust in the decisions taken by deep learning models. In this article, we propose an XAI technique using the fusion of Eigen-CAM and Score-CAM. Eigen-CAM and Score-CAM are gradient-free methods that do not suffer from gradient saturation. Therefore, we propose the Score-Enhanced Eigen-CAM (SE-CAM), the fusion of Score-CAM and Eigen-CAM. In addition to SE-CAM, the research work also trained the VGG16 architecture for multi-class classification of chest X-rays. In addition, we examined SE-CAM on the pretrained DenseNet121 using the ChestXpert dataset. The VGG16 model achieves 94.58% accuracy for the classification of COVID-19, pneumonia, lung opacity, and normal chest X-rays. To evaluate the SE-CAM performance, the % average drop and % increase in confidence were used. For the VGG16 and COVID-19 datasets, the % average drop was 18.41 and the % increase in confidence was 45.03. The SE-CAM model outperformed the base XAI models by 5.73 % on the % average drop and % increase in confidence by 7.7 %. \u003c/p\u003e","manuscriptTitle":"Score-Enhanced Eigen-CAM (SE-CAM) Gradient Free Visual Explanation Method","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-05-18 03:12:46","doi":"10.21203/rs.3.rs-9434509/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2026-05-22T10:42:36+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-05-22T03:10:05+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"278312110195895792604071324289980877342","date":"2026-05-22T02:54:34+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-05-16T17:03:25+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"270197022703045659371495712661054485815","date":"2026-05-08T03:31:48+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-05-08T03:02:30+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2026-04-29T02:31:31+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-04-20T11:08:42+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-04-20T11:08:32+00:00","index":"","fulltext":""},{"type":"submitted","content":"Discover Applied Sciences","date":"2026-04-16T07:10:57+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"discover-applied-sciences","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"Learn more about [Discover Applied Sciences](https://link.springer.com/journal/42452)","snPcode":"42452","submissionUrl":"https://submission.springernature.com/new-submission/42452/3","title":"Discover Applied Sciences","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Discover Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"51ab4867-8318-4e1a-b39e-04fd321f59bf","owner":[],"postedDate":"May 18th, 2026","published":true,"recentEditorialEvents":[{"type":"decision","content":"Revision requested","date":"2026-05-22T10:42:36+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-05-22T03:10:05+00:00","index":33,"fulltext":""},{"type":"reviewerAgreed","content":"278312110195895792604071324289980877342","date":"2026-05-22T02:54:34+00:00","index":32,"fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-05-16T17:03:25+00:00","index":25,"fulltext":""},{"type":"reviewerAgreed","content":"270197022703045659371495712661054485815","date":"2026-05-08T03:31:48+00:00","index":22,"fulltext":""},{"type":"reviewersInvited","content":"14","date":"2026-05-08T03:02:30+00:00","index":"","fulltext":""}],"rejectedJournal":[],"revision":"","amendment":"","status":"in-revision","subjectAreas":[],"tags":[],"updatedAt":"2026-05-22T10:53:58+00:00","versionOfRecord":[],"versionCreatedAt":"2026-05-18 03:12:46","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9434509","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9434509","identity":"rs-9434509","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}