[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"detail-sidebar-cat-0-en-105":3,"doc-seo-156499-105":59,"doc-detail-156499-en":130},{"code":4,"msg":5,"data":6},0,"success",[7,13,18,23,28,33,38,43,48,51,55],{"id":8,"doc_module":4,"doc_module_name":9,"category_name":10,"show_sort_weight":11,"slug":12},1,"Document","Story & Novel",90,"story-novel",{"id":14,"doc_module":4,"doc_module_name":9,"category_name":15,"show_sort_weight":16,"slug":17},2,"Literature",80,"literature",{"id":19,"doc_module":4,"doc_module_name":9,"category_name":20,"show_sort_weight":21,"slug":22},4,"Exam",70,"exam",{"id":24,"doc_module":4,"doc_module_name":9,"category_name":25,"show_sort_weight":26,"slug":27},5,"Comic",60,"comic",{"id":29,"doc_module":4,"doc_module_name":9,"category_name":30,"show_sort_weight":31,"slug":32},6,"Technology",50,"technology",{"id":34,"doc_module":4,"doc_module_name":9,"category_name":35,"show_sort_weight":36,"slug":37},7,"Healthcare",40,"healthcare",{"id":39,"doc_module":4,"doc_module_name":9,"category_name":40,"show_sort_weight":41,"slug":42},8,"Research & Report",30,"research-report",{"id":44,"doc_module":4,"doc_module_name":9,"category_name":45,"show_sort_weight":46,"slug":47},9,"Religion & Spirituality",20,"religion-spirituality",{"id":46,"doc_module":4,"doc_module_name":9,"category_name":49,"show_sort_weight":46,"slug":50},"World Cup","world-cup",{"id":52,"doc_module":4,"doc_module_name":9,"category_name":53,"show_sort_weight":52,"slug":54},10,"Lifestyle","lifestyle",{"id":56,"doc_module":4,"doc_module_name":9,"category_name":57,"show_sort_weight":24,"slug":58},19,"General","general",{"code":4,"msg":60,"data":61},"ok",{"site_id":62,"language":63,"slug":64,"title":65,"keywords":66,"description":67,"schema_data":68,"social_meta":123,"head_meta":125,"extra_data":127,"updated_unix":129},105,"en","bipartite-patient-modality-graph-learning-with-event-conditional-modelling-of-censoring-for-cancer-survival-prediction-paper","Bipartite Patient-Modality Graph Learning with Event-Conditional Modelling of Censoring for Cancer Survival Prediction - Paper","","Accurate cancer survival prediction is critical for personalized treatment, yet prior approaches underutilize censored samples and often degrade when modalities are missing. This study presents CenSurv, a bipartite patient-modality graph learning framework with event-conditional censoring modeling. It learns multimodal representations via graph structure, simulates patient-modality relationships under modality-missing scenarios, and uses complete-incomplete alignment for modality-agnostic features. A plug-and-play ECMC module selects reliable censored data and updates survival times during training. Evaluations on five public cancer datasets show CenSurv exceeds state-of-the-art by 3.1% mean C-index and maintains strong robustness; adding ECMC increases mean C-index by 1.3% across eight baselines.",{"@graph":69,"@context":122},[70,84,105],{"@type":71,"itemListElement":72},"BreadcrumbList",[73,77,79,82],{"item":74,"name":75,"@type":76,"position":8},"https://docshare.wps.com","Home","ListItem",{"item":78,"name":9,"@type":76,"position":14},"https://docshare.wps.com/document/",{"item":80,"name":40,"@type":76,"position":81},"https://docshare.wps.com/document/research-report/",3,{"item":83,"name":65,"@type":76,"position":19},"https://docshare.wps.com/document/bipartite-patient-modality-graph-learning-with-event-conditional-modelling-of-censoring-for-cancer-survival-prediction-paper/156499/",{"url":83,"name":65,"@type":85,"image":86,"author":91,"headline":65,"publisher":94,"fileFormat":97,"inLanguage":63,"description":67,"dateModified":98,"datePublished":99,"encodingFormat":97,"isAccessibleForFree":100,"interactionStatistic":101},"DigitalDocument",{"url":87,"@type":88,"width":89,"height":90},"https://docshare.wps.com/thumbnails/bipartite-patient-modality-graph-learning-with-event-conditional-modelling-of-censoring-for-cancer-survival-prediction-paper/156499.png","ImageObject",300,407,{"name":92,"@type":93},"Aurelia","Person",{"url":74,"name":95,"@type":96},"DocShare","Organization","application/pdf","2026-10-16","2026-08-28",true,{"@type":102,"interactionType":103,"userInteractionCount":52},"InteractionCounter",{"@type":104},"ViewAction",{"@type":106,"mainEntity":107},"FAQPage",[108,114,118],{"name":109,"@type":110,"acceptedAnswer":111},"What problem does CenSurv address in cancer survival prediction?","Question",{"text":112,"@type":113},"CenSurv targets two issues: censored samples are often ignored or underutilized, and performance can drop in modality-missing scenarios during both training and inference.","Answer",{"name":115,"@type":110,"acceptedAnswer":116},"How does CenSurv model patient-modality relationships?",{"text":117,"@type":113},"It uses bipartite graph learning to simulate the patient-modality relationship under different modality-missing scenarios and to learn modality-agnostic features via complete-incomplete alignment.",{"name":119,"@type":110,"acceptedAnswer":120},"What is the role of the plug-and-play ECMC module?",{"text":121,"@type":113},"ECMC performs event-conditional modeling of censoring by selecting reliable censored data using dynamic confidence accumulation, assigning more accurate survival times, and incorporating the updated data as uncensored during training.","https://schema.org",{"og:url":83,"og:type":124,"og:title":65,"og:site_name":95,"og:description":67},"article",{"robots":126,"canonical":83},"index,follow",{"doc_id":128,"site_id":62},156499,1787960995,{"code":4,"msg":5,"data":131},{"doc_id":128,"user_id":132,"nickname":92,"user_avatar":133,"doc_module":4,"category_id":39,"category_name":40,"doc_title":65,"doc_description":67,"doc_content":134,"file_id":135,"file_url":136,"file_type":137,"file_size":138,"view_count":52,"is_deleted":4,"is_public":8,"is_downloadable":8,"audit_status":8,"page_count":139,"language":140,"language_code":63,"site_id":62,"html_lang":63,"table_of_contents":141,"faqs":142,"seo_title":143,"seo_description":67,"update_tm":129,"read_time":144,"preview_page_count":4},1099514068365,"https://ap-avatar.wpscdn.com/avatar/10000253d8d9f28188e?_k=1776742907772140068","Bipartite Patient-Modality Graph Learning with Event-Conditional Modelling of Censoring for Cancer Survival Prediction  \nHailin Yue 1 , Hulin Kuang 1⋆, Jin Liu 1 ,2 , Junjian Li 1 , Lanlan Wang 1 , Mengshen  \nHe 1 , and Jianxin Wang 1 ,2  \n1 Hunan Provincial Key Lab on Bioinformatics, School of Computer Science and Engineering, Central South University, Changsha, China  \n2Xinjiang Engineering Research Center of Big Data and Intelligent Software, School of Software, Xinjiang University, Urumqi, China  \n[hulinkuang@csu.edu.cn](hulinkuang@csu.edu.cn)  \nAbstract. Accurately predicting the survival of cancer patients is crucial for personalized treatment. However, existing studies focus solely on the relationships between samples with known survival risks, without fully leveraging the value of censored samples. Furthermore, these studies may suffer performance degradation in modality-missing scenariosand even struggle during the inference process. In this study, we propose a bipartite patient-modality graph learning with event-conditional modelling of censoring for cancer survival prediction (CenSurv) . Specifically, we first use graph structure to model multimodal data and obtain representation. Then, to alleviate performance degradation in modalitymissing scenarios, we design a bipartite graph to simulate the patientmodality relationship in various modality-missing scenarios and leverage a complete-incomplete alignment strategy to explore modality-agnostic features. Finally, we design a plug-and-play event-conditional modeling of censoring (ECMC) that selects reliable censored data using dynamic momentum accumulation confidences, assigns more accurate survival times to these censored data, and incorporates them as uncensored data into training. Comprehensive evaluations on 5 publicly cancer datasets showcase the superiority of CenSurv over the best state-of-theart by 3 . 1% in terms of the mean C-index, while also exhibiting excellent robustness under various modality-missing scenarios. In addition, using the plug-and-play ECMC module, the mean C-index of 8 baselines increased by 1 .3% across 5 datasets. Code of CenSurv is available at [https://github.com/yuehailin/CenSurv](https://github.com/yuehailin/CenSurv).  \nKeywords: Survival prediction · Censored data · Plug-and-play.  \n1 Introduction  \nGlobally, cancer is a huge public health challenge. According to data from the World Health Organization (WHO), millions of people are diagnosed with cancer  \n⋆ Corresponding author  \n2 H. Yue et al.  \neach year, leading to millions of deaths [24] . Personalized treatment for patients is recommended in clinical practice [14,11] . To guide personalized treatment for patients, it is necessary to accurately predict the survival of cancer patients.  \nExamination Treatment  \nFollow up  \nDead/Alive  \n(a)  \nAlive  \nUncensored data  \n􀜶􀝅􀝉􀝁 = 􀜶􀮽􀯘􀯔􀯗 − 􀜶􀯌􀯧􀯔􀯥􀯧 Statue : 􀟜 = 1  \nCensored data  \n􀜶􀝅􀝉􀝁 = 􀜶􀮿􀯢􀯟􀯟􀯢􀯪−􀯨􀯣 − 􀜶􀯌􀯧􀯔􀯥􀯧 Statue : 􀟜 = 0  \n(b)  \nFig. 1. (a) The distinction of censored and uncensored data. For uncensored data, the survival status is dead (δ = 0), and the survival time is accurate. For censored data, the survival status is alive (δ=1), and the survival time calculated with the follow-up time as the event endpoint is inaccurate and smaller than the actual survival time. (b) The number of censored and uncensored data on 5 datasets in our study.  \nWith advances in medical imaging and genomics, many studies have identified potential markers for cancer survival prediction [20,5] . For example, Shao et al. proposed a hierarchical vision transformer for survival prediction using whole slide images [20] . Increasingly, studies combined imaging, genetic, and clinical data for more accurate survival prediction [8,12] . For example, Chen et al. used both pathological images and genetic data for survival prediction [2] . Other methods like MultiSurv [26], SurvMamba [4] and SurvPath [9] also used these multimodal data for survival prediction. However, th","cbCaiuu525nqEtX2","https://ap.wps.com/l/cbCaiuu525nqEtX2","pdf",1516332,11,"English","# Introduction\n## Problem background: cancer survival prediction and censored data\n## Limitations of existing methods in modality-missing scenarios\n## Proposed approach: CenSurv framework\n## Main contributions","[{\"question\":\"What problem does CenSurv address in cancer survival prediction?\",\"answer\":\"CenSurv targets two issues: censored samples are often ignored or underutilized, and performance can drop in modality-missing scenarios during both training and inference.\"},{\"question\":\"How does CenSurv model patient-modality relationships?\",\"answer\":\"It uses bipartite graph learning to simulate the patient-modality relationship under different modality-missing scenarios and to learn modality-agnostic features via complete-incomplete alignment.\"},{\"question\":\"What is the role of the plug-and-play ECMC module?\",\"answer\":\"ECMC performs event-conditional modeling of censoring by selecting reliable censored data using dynamic confidence accumulation, assigning more accurate survival times, and incorporating the updated data as uncensored during training.\"}]","Bipartite Patient-Modality Graph Learning with Event-Conditional Modelling of Censoring for Cancer Survival Prediction - Paper | PDF",28]