Explainable Attention-Enhanced Heuristic Paradigm for Multi-View Prognostic Risk Sore Development in Hepatocellular Carcinoma

preprint OA: closed CC-BY-4.0
📄 Open PDF Full text JSON View at publisher

Abstract

Abstract Purpose Existing prognostic staging systems depend on expensive manual extraction by pathologists, potentially overlooking hidden information, or use black-box deep learning models, which limits their clinical acceptance.This study introduces a novel deep learning-assisted paradigm for creating interpretable, multi-view risk scores to stratify prognostic risk in hepatocellular carcinoma (HCC) patients. Methods 510 HCC patients were enrolled in an internal dataset (SYSUCC) as training and validation cohorts to develop the Hybrid Deep Score (HDS): The Attention Activator (ATAT) was designed to heuristically identify tissues associated with high prognostic risk, and a multi-view risk scoring system based on ATAT established HDS from microscopic to macroscopic levels. The HDS was also validated on an external testing cohort (TCGA-LIHC) with 341 HCC patients. We assessed the prognostic significance using Cox regression and the concordance index (c-index). Results The ATAT first heuristically identified regions where necrosis, lymphocytes, and tumor tissues converge, particularly focusing on their junctions in high-risk patients. From this, this study developed three independent risk factors: microscopic morphological, co-localization, and deep global indicators, ultimately predicting HDS for each patient. The HDS outperformed existing clinical prognostic staging systems, showing higher hazard ratios (HR 3.24, 95% CI 1.91-5.43 in SYSUCC; HR 2.34, 95% CI 1.58-3.47 in TCGA-LIHC) and c-index (0.751 in SYSUCC; 0.729 in TCGA-LIHC) for Disease-Free Survival (DFS). Conclusion This novel paradigm, from identifying high-risk tissues to constructing prognostic risk scores, offers fresh insights into HCC research. It more precisely stratifies HCC patients into high- and low-risk groups for DFS and Overall Survival (OS) compared to existing clinical risk staging systems.
Full text 19,232 characters · extracted from preprint-html · click to expand
Explainable Attention-Enhanced Heuristic Paradigm for Multi-View Prognostic Risk Sore Development in Hepatocellular Carcinoma | 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 Explainable Attention-Enhanced Heuristic Paradigm for Multi-View Prognostic Risk Sore Development in Hepatocellular Carcinoma Anran Liu, Jiang Zhang, Tong Li, Danyang Zheng, Yihong Ling, Lianghe Lu, and 2 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-5480986/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 16 Mar, 2025 Read the published version in Hepatology International → Version 1 posted 4 You are reading this latest preprint version Abstract Purpose Existing prognostic staging systems depend on expensive manual extraction by pathologists, potentially overlooking hidden information, or use black-box deep learning models, which limits their clinical acceptance.This study introduces a novel deep learning-assisted paradigm for creating interpretable, multi-view risk scores to stratify prognostic risk in hepatocellular carcinoma (HCC) patients. Methods 510 HCC patients were enrolled in an internal dataset (SYSUCC) as training and validation cohorts to develop the Hybrid Deep Score (HDS): The Attention Activator (ATAT) was designed to heuristically identify tissues associated with high prognostic risk, and a multi-view risk scoring system based on ATAT established HDS from microscopic to macroscopic levels. The HDS was also validated on an external testing cohort (TCGA-LIHC) with 341 HCC patients. We assessed the prognostic significance using Cox regression and the concordance index (c-index). Results The ATAT first heuristically identified regions where necrosis, lymphocytes, and tumor tissues converge, particularly focusing on their junctions in high-risk patients. From this, this study developed three independent risk factors: microscopic morphological, co-localization, and deep global indicators, ultimately predicting HDS for each patient. The HDS outperformed existing clinical prognostic staging systems, showing higher hazard ratios (HR 3.24, 95% CI 1.91-5.43 in SYSUCC; HR 2.34, 95% CI 1.58-3.47 in TCGA-LIHC) and c-index (0.751 in SYSUCC; 0.729 in TCGA-LIHC) for Disease-Free Survival (DFS). Conclusion This novel paradigm, from identifying high-risk tissues to constructing prognostic risk scores, offers fresh insights into HCC research. It more precisely stratifies HCC patients into high- and low-risk groups for DFS and Overall Survival (OS) compared to existing clinical risk staging systems. Hepatocellular carcinoma Prognostic risk scoring Deep learning-assisted diagnosis Whole slide images Necrosis Lymphocytes Interpretability Disease-free survival Overall survival Artificial intelligence Full Text Supplementary Files DemographicsClassification4.pdf DemographicsInhouse3.pdf DemographicsTCGA6.pdf EvaluationClassificationATAT8.pdf ForestTCGA9.pdf UnivariateMultivariateSYSUCC5.pdf cindexAUCinhouse1.pdf ATATMultiViewRiskScoringSystem7.pdf cindexAUCtcga2.pdf Cite Share Download PDF Status: Published Journal Publication published 16 Mar, 2025 Read the published version in Hepatology International → Version 1 posted Reviewers agreed at journal 24 Nov, 2024 Reviewers invited by journal 23 Nov, 2024 Editor assigned by journal 20 Nov, 2024 First submitted to journal 18 Nov, 2024 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-5480986","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":381825679,"identity":"07ab9a2e-6908-4317-9b72-31fd39572796","order_by":0,"name":"Anran Liu","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABAklEQVRIiWNgGAWjYPACCx4G9gYk/gMgNuDBq0WCh4HnAIQJVphAhBYgSiBSC//s3oefCxgkZAxuPn74uHCHjbw9A+/BB4lthxnMYfaiG3/nuLH0DKDDDG6nGRvPPJNm2MPAl2wA0mLZ24BVC8ONNAZpHrCWHDZp3rbDCUBfmUmAtBicx65D/kYa82+wlptnQFr+E9ZicCONDWLLDR6QlgNIWs5id5ghUIs1j4EEj+QZoF9425INew4D/ZJwLp3H4Ax278sBHXabp8LGnu/44YePedvs5Nnbew8++FBmLWdwJgG7yyDOQ+YwA+ODkY0Bf0SiAZDiP6RoGAWjYBSMgmEOACYHT3F5XbXYAAAAAElFTkSuQmCC","orcid":"https://orcid.org/0000-0003-0213-2923","institution":"The Hong Kong Polytechnic University","correspondingAuthor":true,"prefix":"","firstName":"Anran","middleName":"","lastName":"Liu","suffix":""},{"id":381825680,"identity":"58a73ba1-b229-4115-86b2-dbf20ed556bb","order_by":1,"name":"Jiang Zhang","email":"","orcid":"","institution":"The Hong Kong Polytechnic University","correspondingAuthor":false,"prefix":"","firstName":"Jiang","middleName":"","lastName":"Zhang","suffix":""},{"id":381825681,"identity":"9ee3e1ef-7de5-4457-8132-0b25264b9164","order_by":2,"name":"Tong Li","email":"","orcid":"","institution":"Washington University In St Louis: Washington University in St Louis","correspondingAuthor":false,"prefix":"","firstName":"Tong","middleName":"","lastName":"Li","suffix":""},{"id":381825682,"identity":"bf94201f-60c0-42d8-8c1f-6e775df6e388","order_by":3,"name":"Danyang Zheng","email":"","orcid":"","institution":"Sun Yat-sen University First Affiliated Hospital","correspondingAuthor":false,"prefix":"","firstName":"Danyang","middleName":"","lastName":"Zheng","suffix":""},{"id":381825683,"identity":"c38a683d-ff60-45e9-98f7-680de76b887a","order_by":4,"name":"Yihong Ling","email":"","orcid":"","institution":"Sun Yat-sen University State Key Laboratory of Oncology in South China","correspondingAuthor":false,"prefix":"","firstName":"Yihong","middleName":"","lastName":"Ling","suffix":""},{"id":381825684,"identity":"4a1baf51-9ab9-4f7b-a496-2e70b537e29c","order_by":5,"name":"Lianghe Lu","email":"","orcid":"","institution":"Sun Yat-sen University State Key Laboratory of Oncology in South China","correspondingAuthor":false,"prefix":"","firstName":"Lianghe","middleName":"","lastName":"Lu","suffix":""},{"id":381825685,"identity":"385ec374-c2b1-4a03-b7a9-46575fa339c6","order_by":6,"name":"Yuanpeng Zhang","email":"","orcid":"","institution":"The Hong Kong Polytechnic University","correspondingAuthor":false,"prefix":"","firstName":"Yuanpeng","middleName":"","lastName":"Zhang","suffix":""},{"id":381825686,"identity":"f139b4bc-afae-43bf-9502-7034da5edf2e","order_by":7,"name":"Jing Cai","email":"","orcid":"","institution":"The Hong Kong Polytechnic University","correspondingAuthor":false,"prefix":"","firstName":"Jing","middleName":"","lastName":"Cai","suffix":""}],"badges":[],"createdAt":"2024-11-19 07:11:57","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-5480986/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-5480986/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1007/s12072-025-10793-8","type":"published","date":"2025-03-16T15:57:08+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":78688908,"identity":"7223ba4c-6d20-4874-a444-44cd1a501ce7","added_by":"auto","created_at":"2025-03-17 16:06:44","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2000502,"visible":true,"origin":"","legend":"","description":"","filename":"HEPID2401565.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5480986/v1_covered_afb3d2ae-f581-4197-9f91-25037814ffef.pdf"},{"id":70592355,"identity":"40c1f3b5-6b46-4dd9-ba9f-223d2b8bc0b4","added_by":"auto","created_at":"2024-12-04 17:22:37","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":114510,"visible":true,"origin":"","legend":"","description":"","filename":"DemographicsClassification4.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5480986/v1/02206b42e99f770178d7200c.pdf"},{"id":70592836,"identity":"0c4261b9-47c5-4311-8cc0-8b16cbd5975c","added_by":"auto","created_at":"2024-12-04 17:30:38","extension":"pdf","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":143644,"visible":true,"origin":"","legend":"","description":"","filename":"DemographicsInhouse3.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5480986/v1/9e1d54364df55df14d3d2d12.pdf"},{"id":70592356,"identity":"fca1fd18-1863-4c23-a5fd-e41d5bc955ab","added_by":"auto","created_at":"2024-12-04 17:22:38","extension":"pdf","order_by":3,"title":"","display":"","copyAsset":false,"role":"supplement","size":115130,"visible":true,"origin":"","legend":"","description":"","filename":"DemographicsTCGA6.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5480986/v1/6201583e2ab7de8413a7c405.pdf"},{"id":70592362,"identity":"db9b1c08-02a0-4f67-a55d-6312730b060c","added_by":"auto","created_at":"2024-12-04 17:22:38","extension":"pdf","order_by":4,"title":"","display":"","copyAsset":false,"role":"supplement","size":284478,"visible":true,"origin":"","legend":"","description":"","filename":"EvaluationClassificationATAT8.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5480986/v1/209ba16320fdb4036130aa80.pdf"},{"id":70592363,"identity":"5187275c-5f43-4c7a-af22-830021a855bc","added_by":"auto","created_at":"2024-12-04 17:22:38","extension":"pdf","order_by":5,"title":"","display":"","copyAsset":false,"role":"supplement","size":242103,"visible":true,"origin":"","legend":"","description":"","filename":"ForestTCGA9.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5480986/v1/4dc52a14792c68601d1be757.pdf"},{"id":70592359,"identity":"3942221d-4fe1-4139-992b-81be8f161ba4","added_by":"auto","created_at":"2024-12-04 17:22:38","extension":"pdf","order_by":6,"title":"","display":"","copyAsset":false,"role":"supplement","size":155502,"visible":true,"origin":"","legend":"","description":"","filename":"UnivariateMultivariateSYSUCC5.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5480986/v1/2616b0052f33194eb1b91047.pdf"},{"id":70592837,"identity":"8b888ec2-df24-4542-9d85-59ba9c37e176","added_by":"auto","created_at":"2024-12-04 17:30:38","extension":"pdf","order_by":7,"title":"","display":"","copyAsset":false,"role":"supplement","size":137845,"visible":true,"origin":"","legend":"","description":"","filename":"cindexAUCinhouse1.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5480986/v1/f2430e487261ad42a0e72282.pdf"},{"id":70592360,"identity":"a2ef2782-7b53-4723-9fe9-b32a004661c8","added_by":"auto","created_at":"2024-12-04 17:22:38","extension":"pdf","order_by":8,"title":"","display":"","copyAsset":false,"role":"supplement","size":193996,"visible":true,"origin":"","legend":"","description":"","filename":"ATATMultiViewRiskScoringSystem7.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5480986/v1/f0cb8eb00c2708c599d78f11.pdf"},{"id":70592839,"identity":"ebb3236d-2a2b-4f23-a0f1-dde1d5050996","added_by":"auto","created_at":"2024-12-04 17:30:38","extension":"pdf","order_by":9,"title":"","display":"","copyAsset":false,"role":"supplement","size":155563,"visible":true,"origin":"","legend":"","description":"","filename":"cindexAUCtcga2.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5480986/v1/67a54ed135ba94b5b503ef35.pdf"}],"financialInterests":"","formattedTitle":"Explainable Attention-Enhanced Heuristic Paradigm for Multi-View Prognostic Risk Sore Development in Hepatocellular Carcinoma","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":true,"isPdf":true,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"hepatology-international","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"hepi","sideBox":"Learn more about [Hepatology International](https://www.springer.com/journal/12072)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/hepi/default.aspx","title":"Hepatology International","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"Hepatocellular carcinoma , Prognostic risk scoring , Deep learning-assisted diagnosis , Whole slide images , Necrosis , Lymphocytes , Interpretability , Disease-free survival , Overall survival , Artificial intelligence","lastPublishedDoi":"10.21203/rs.3.rs-5480986/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-5480986/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cstrong\u003ePurpose\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eExisting prognostic staging systems depend on expensive manual extraction by pathologists, potentially overlooking hidden information, or use black-box deep learning models, which limits their clinical acceptance.This study introduces a novel deep learning-assisted paradigm for creating interpretable, multi-view risk scores to stratify prognostic risk in hepatocellular carcinoma (HCC) patients.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMethods\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e510 HCC patients were enrolled in an internal dataset (SYSUCC) as training and validation cohorts to develop the Hybrid Deep Score (HDS): The Attention Activator (ATAT) was designed to heuristically identify tissues associated with high prognostic risk, and a multi-view risk scoring system based on ATAT established HDS from microscopic to macroscopic levels. The HDS was also validated on an external testing cohort (TCGA-LIHC) with 341 HCC patients. We assessed the prognostic significance using Cox regression and the concordance index (c-index).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eResults\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe ATAT first heuristically identified regions where necrosis, lymphocytes, and tumor tissues converge, particularly focusing on their junctions in high-risk patients. From this, this study developed three independent risk factors: microscopic morphological, co-localization, and deep global indicators, ultimately predicting HDS for each patient. The HDS outperformed existing clinical prognostic staging systems, showing higher hazard ratios (HR 3.24, 95% CI 1.91-5.43 in SYSUCC; HR 2.34, 95% CI 1.58-3.47 in TCGA-LIHC) and c-index (0.751 in SYSUCC; 0.729 in TCGA-LIHC) for Disease-Free Survival (DFS).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConclusion\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis novel paradigm, from identifying high-risk tissues to constructing prognostic risk scores, offers fresh insights into HCC research. It more precisely stratifies HCC patients into high- and low-risk groups for DFS and Overall Survival (OS) compared to existing clinical risk staging systems.\u003c/p\u003e","manuscriptTitle":"Explainable Attention-Enhanced Heuristic Paradigm for Multi-View Prognostic Risk Sore Development in Hepatocellular Carcinoma","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-12-04 17:22:33","doi":"10.21203/rs.3.rs-5480986/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"reviewerAgreed","content":"","date":"2024-11-24T10:34:50+00:00","index":0,"fulltext":""},{"type":"reviewersInvited","content":"","date":"2024-11-24T02:38:13+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2024-11-20T16:07:14+00:00","index":"","fulltext":""},{"type":"submitted","content":"Hepatology International","date":"2024-11-19T02:10:46+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"hepatology-international","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"hepi","sideBox":"Learn more about [Hepatology International](https://www.springer.com/journal/12072)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/hepi/default.aspx","title":"Hepatology International","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"d65adb3c-c80d-484e-95e9-b2900da024fa","owner":[],"postedDate":"December 4th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2025-03-17T15:59:53+00:00","versionOfRecord":{"articleIdentity":"rs-5480986","link":"https://doi.org/10.1007/s12072-025-10793-8","journal":{"identity":"hepatology-international","isVorOnly":false,"title":"Hepatology International"},"publishedOn":"2025-03-16 15:57:08","publishedOnDateReadable":"March 16th, 2025"},"versionCreatedAt":"2024-12-04 17:22:33","video":"","vorDoi":"10.1007/s12072-025-10793-8","vorDoiUrl":"https://doi.org/10.1007/s12072-025-10793-8","workflowStages":[]},"version":"v1","identity":"rs-5480986","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-5480986","identity":"rs-5480986","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}

Text is read by the "Ask this paper" AI Q&A widget below. Extraction quality varies by source — PMC NXML preserves structure cleanly, OA-HTML may include some navigation residue, and OA-PDF can have broken hyphenation. The publisher copy (via DOI) is the canonical version.

My notes (saved in your browser only)

Ask this paper AI returns verbatim quotes from the full text · source: preprint-html

Answers must be backed by verbatim quotes from this paper's full text. Hallucinated quotes are dropped automatically; if no verbatim passage answers the question, we say so. How this works

Citation neighborhood (no data yet)

We don't have any in-corpus citations linked to this paper yet. This is a recent paper (2024) — citers typically take a year or two to land, and the OpenAlex reference graph may still be filling in.

Source provenance

europepmc
last seen: 2026-05-20T01:45:00.602351+00:00
unpaywall
last seen: 2026-05-27T02:00:06.600101+00:00
License: CC-BY-4.0