Feedback-Driven Cascaded Framework for Medical Image Detection and Segmentation

Feedback-Driven Cascaded Framework for Medical Image Detection and Segmentation | 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 Feedback-Driven Cascaded Framework for Medical Image Detection and Segmentation Md Waheduzzaman Tuhin, Tirtha Raj Kandel, Rajneesh Kumar Singh, and 4 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9446374/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 12 You are reading this latest preprint version Abstract Image diagnosis Medical image diagnosis [1] is a technique that is important in the contemporary healthcare delivery systems since it directly aids in clinical decision-making, disease diagnosis, and treatment plans. Nevertheless, such a task is linked to a variety of issues, including differences in anatomical morphology, noise and artifact, and imbalance in the classes, and it is challenging to obtain high diagnostic accuracy in different imaging modalities. The current deep learning models tend to lose their consistency in performance when they are used to heterogeneous medical datasets, therefore restricting their generalizability and robustness [2]. In this paper, we introduce a new hybrid model of image detection and segmentation of medical images that exploits a cascaded architecture with an iterative feedback-based learning process. The system proposed is a combination of various models organized in a series where each model is an improvement of the results of the other model, and in this way, the quality of the segmentation gets improved. Moreover, a dynamic backward feedback mechanism is proposed, allowing subsequent stage models to inform and calibrate previous stage representations, which helps minimize error propagation and learn features better. All the models in the framework are structured around an Enhanced Feedback U-Net architecture, with attention gates, residual connections and a discrete feedback fusion module to combine adaptive feature integration. Training of the system is done with a multi-objective loss function that ensures there is accuracy, consistency and gradual refinement and improvement throughout the entire training. Experimental findings indicate that the suggested method outperforms the traditional methods in regard to accuracy in segmentation, and robustness, as well as the accuracy of the boundaries. The model offers a credible and flexible system to analyse medical images, and has good possibilities of implementation into clinical practice. Medical Image Detection Hybrid Model Cascading Network Backward Feedback U-Net Attention Gates Deep Learning Segmentation Diagnostics Artificial Intelligence Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Review Version 1 posted Reviews received at journal 13 May, 2026 Reviews received at journal 10 May, 2026 Reviewers agreed at journal 08 May, 2026 Reviewers agreed at journal 07 May, 2026 Reviewers agreed at journal 07 May, 2026 Reviewers agreed at journal 07 May, 2026 Reviews received at journal 07 May, 2026 Reviewers agreed at journal 07 May, 2026 Reviewers invited by journal 07 May, 2026 Editor assigned by journal 24 Apr, 2026 Submission checks completed at journal 24 Apr, 2026 First submitted to journal 17 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-9446374","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":639622014,"identity":"360d3289-8dfe-422e-b267-ffac665bcb1e","order_by":0,"name":"Md Waheduzzaman Tuhin","email":"","orcid":"","institution":"Sharda University","correspondingAuthor":false,"prefix":"","firstName":"Md","middleName":"Waheduzzaman","lastName":"Tuhin","suffix":""},{"id":639622016,"identity":"c3aaba9c-78dc-4dbb-bc9e-8571d2135050","order_by":1,"name":"Tirtha Raj Kandel","email":"","orcid":"","institution":"Sharda University","correspondingAuthor":false,"prefix":"","firstName":"Tirtha","middleName":"Raj","lastName":"Kandel","suffix":""},{"id":639622020,"identity":"e8b0eda7-e51e-4616-8b43-9ae7b48c6d94","order_by":2,"name":"Rajneesh Kumar Singh","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABB0lEQVRIiWNgGAWjYBACCQTzAJiQA5MPSNFiDCYTiNMC0ZfYAKLwaZGckXtM4ucehnz+xrMHH3z4cyd9ftjhh0Bb7OR0G7BrkZbIS5PsecZgOePAuWTDmW3PcjfeTjMAakk2NjuAXYucRI7ZDZ4DDAYMB86YSfM2HM7dODsBpOVA4jY8Wm7+AWqRP3DG/PefP4fTDWenf8CrRRqo5TbIFgOgLcwMbIcT5KVz8Nsi2fPG/LfMAQkDwwNnjCV72w4bbpDOKTiQYIDbLxLHc4wN3xywMZC7ccbww48/h+XlZ6dv/vChwk4OlxaYTiCCqjAA0wZ4lUMBfwOElm8gRvUoGAWjYBSMJAAAHnFoi0JsKKIAAAAASUVORK5CYII=","orcid":"","institution":"Sharda University","correspondingAuthor":true,"prefix":"","firstName":"Rajneesh","middleName":"Kumar","lastName":"Singh","suffix":""},{"id":639622021,"identity":"2b1a64ea-18be-41d1-929c-c33c44162e9e","order_by":3,"name":"Ravi Prakash Chaturvedi","email":"","orcid":"","institution":"Sharda University","correspondingAuthor":false,"prefix":"","firstName":"Ravi","middleName":"Prakash","lastName":"Chaturvedi","suffix":""},{"id":639622022,"identity":"e193c029-354f-4573-b413-9247694bd6d7","order_by":4,"name":"Annu Mishra","email":"","orcid":"","institution":"Sharda University","correspondingAuthor":false,"prefix":"","firstName":"Annu","middleName":"","lastName":"Mishra","suffix":""},{"id":639622023,"identity":"1af469d9-4449-424d-95f7-223330f01a14","order_by":5,"name":"Anurag Rai","email":"","orcid":"","institution":"Sharda University","correspondingAuthor":false,"prefix":"","firstName":"Anurag","middleName":"","lastName":"Rai","suffix":""},{"id":639622024,"identity":"3a5d3662-9512-4c15-8d9b-221d6d97816f","order_by":6,"name":"Aditya Dayal Tyagi","email":"","orcid":"","institution":"Sharda University","correspondingAuthor":false,"prefix":"","firstName":"Aditya","middleName":"Dayal","lastName":"Tyagi","suffix":""}],"badges":[],"createdAt":"2026-04-17 08:38:44","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-9446374/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9446374/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":109759315,"identity":"ab63a45b-96d4-40c6-942b-2d63e186340a","added_by":"auto","created_at":"2026-05-22 07:26:37","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":659036,"visible":true,"origin":"","legend":"","description":"","filename":"FeedbackDrivenCascadedFrameworkforMedicalImageDetectionandSegmentationv2withdeclarations.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9446374/v1_covered_8a1e0478-3f91-4cf7-ac53-160b5cba1294.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Feedback-Driven Cascaded Framework for Medical Image Detection and Segmentation","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-computing","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"Learn more about [Discover Computing](https://link.springer.com/journal/10791)","snPcode":"10791","submissionUrl":"https://submission.springernature.com/new-submission/10791/3","title":"Discover Computing","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Discover Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Medical Image Detection, Hybrid Model, Cascading Network, Backward Feedback, U-Net, Attention Gates, Deep Learning, Segmentation, Diagnostics, Artificial Intelligence","lastPublishedDoi":"10.21203/rs.3.rs-9446374/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9446374/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eImage diagnosis Medical image diagnosis [1] is a technique that is important in the contemporary healthcare delivery systems since it directly aids in clinical decision-making, disease diagnosis, and treatment plans. Nevertheless, such a task is linked to a variety of issues, including differences in anatomical morphology, noise and artifact, and imbalance in the classes, and it is challenging to obtain high diagnostic accuracy in different imaging modalities. The current deep learning models tend to lose their consistency in performance when they are used to heterogeneous medical datasets, therefore restricting their generalizability and robustness [2]. In this paper, we introduce a new hybrid model of image detection and segmentation of medical images that exploits a cascaded architecture with an iterative feedback-based learning process. The system proposed is a combination of various models organized in a series where each model is an improvement of the results of the other model, and in this way, the quality of the segmentation gets improved. Moreover, a dynamic backward feedback mechanism is proposed, allowing subsequent stage models to inform and calibrate previous stage representations, which helps minimize error propagation and learn features better. All the models in the framework are structured around an Enhanced Feedback U-Net architecture, with attention gates, residual connections and a discrete feedback fusion module to combine adaptive feature integration. Training of the system is done with a multi-objective loss function that ensures there is accuracy, consistency and gradual refinement and improvement throughout the entire training. Experimental findings indicate that the suggested method outperforms the traditional methods in regard to accuracy in segmentation, and robustness, as well as the accuracy of the boundaries. The model offers a credible and flexible system to analyse medical images, and has good possibilities of implementation into clinical practice.\u003c/p\u003e","manuscriptTitle":"Feedback-Driven Cascaded Framework for Medical Image Detection and Segmentation","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-05-18 08:36:32","doi":"10.21203/rs.3.rs-9446374/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"editorInvitedReview","content":"","date":"2026-05-13T09:57:39+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-05-10T10:38:05+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"36946163754152308201585665003060045420","date":"2026-05-08T13:08:59+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"298967045966492680914372936853683364127","date":"2026-05-08T03:42:48+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"297695737564211330706414575205697719170","date":"2026-05-08T00:39:32+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"319143012568234605479291041668999362077","date":"2026-05-08T00:00:43+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-05-07T21:40:06+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"330421496837622952987789595397635040288","date":"2026-05-07T21:33:03+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-05-07T18:02:01+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-04-24T15:16:36+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-04-24T15:16:24+00:00","index":"","fulltext":""},{"type":"submitted","content":"Discover Computing","date":"2026-04-17T08:32:26+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"discover-computing","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"Learn more about [Discover Computing](https://link.springer.com/journal/10791)","snPcode":"10791","submissionUrl":"https://submission.springernature.com/new-submission/10791/3","title":"Discover Computing","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Discover Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"40b24c21-dfbd-4bfb-91b0-9cc3f78f09e1","owner":[],"postedDate":"May 18th, 2026","published":true,"recentEditorialEvents":[{"type":"editorInvitedReview","content":"","date":"2026-05-13T09:57:39+00:00","index":28,"fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-05-10T10:38:05+00:00","index":27,"fulltext":""},{"type":"reviewerAgreed","content":"36946163754152308201585665003060045420","date":"2026-05-08T13:08:59+00:00","index":26,"fulltext":""},{"type":"reviewerAgreed","content":"298967045966492680914372936853683364127","date":"2026-05-08T03:42:48+00:00","index":25,"fulltext":""},{"type":"reviewerAgreed","content":"297695737564211330706414575205697719170","date":"2026-05-08T00:39:32+00:00","index":24,"fulltext":""},{"type":"reviewerAgreed","content":"319143012568234605479291041668999362077","date":"2026-05-08T00:00:43+00:00","index":23,"fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-05-07T21:40:06+00:00","index":22,"fulltext":""},{"type":"reviewerAgreed","content":"330421496837622952987789595397635040288","date":"2026-05-07T21:33:03+00:00","index":21,"fulltext":""},{"type":"reviewersInvited","content":"11","date":"2026-05-07T18:02:01+00:00","index":"","fulltext":""}],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2026-05-18T08:36:32+00:00","versionOfRecord":[],"versionCreatedAt":"2026-05-18 08:36:32","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9446374","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9446374","identity":"rs-9446374","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}