[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"detail-sidebar-cat-0-en-105":3,"doc-seo-352696-105":59,"doc-detail-352696-en":131},{"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":124,"head_meta":126,"extra_data":128,"updated_unix":130},105,"en","empirical-evaluation-of-variability-and-multi-institutional-generalizability-of-deep-learning-survival-models-application-to-renal-cancer-ct-scans","Empirical evaluation of variability and multi-institutional generalizability of deep learning survival models - application to renal cancer CT scans","","Deep learning survival (DLS) modeling shows promise for predicting patient outcomes from biomedical data, yet its methodological drivers for imaging-based generalization remain insufficiently studied. This work systematically tests how data partitioning, data order, model initialization, and augmentation strategies affect CT-based DLS performance when predicting continuous survival outcomes in external validation cohorts. A multi-institutional CT dataset is assembled from public repositories, and a 3D ResNet-18 Cox-loss model is evaluated with concordance index and hazard ratio. Results show covariate-balanced partitioning improves external validation, initialization alters performance variability, and augmentation—especially additive Gaussian noise—consistently enhances generalization.",{"@graph":69,"@context":123},[70,84,106],{"@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/empirical-evaluation-of-variability-and-multi-institutional-generalizability-of-deep-learning-survival-models-application-to-renal-cancer-ct-scans/352696/",{"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/empirical-evaluation-of-variability-and-multi-institutional-generalizability-of-deep-learning-survival-models-application-to-renal-cancer-ct-scans/352696.png","ImageObject",300,407,{"name":92,"@type":93},"Maya Linwood","Person",{"url":74,"name":95,"@type":96},"DocShare","Organization","application/pdf","2026-10-25","2026-09-22",true,{"@type":102,"interactionType":103,"userInteractionCount":105},"InteractionCounter",{"@type":104},"ViewAction",13,{"@type":107,"mainEntity":108},"FAQPage",[109,115,119],{"name":110,"@type":111,"acceptedAnswer":112},"What problem does the study address regarding deep learning survival (DLS) models?","Question",{"text":113,"@type":114},"It examines which methodological choices improve model generalization for CT-based DLS, particularly for continuous survival outcome prediction in external validation cohorts.","Answer",{"name":116,"@type":111,"acceptedAnswer":117},"Which factors did the study systematically evaluate for generalization?",{"text":118,"@type":114},"Data partitioning, data order, model initialization, and image augmentation strategies were tested using controlled experiments.",{"name":120,"@type":111,"acceptedAnswer":121},"What were the key findings on external validation performance?",{"text":122,"@type":114},"Covariate-balanced partitioning outperformed random partitioning, model initialization significantly affected average performance and variance, and combining augmentations improved external validation, with additive Gaussian noise being the most impactful individual augmentation.","https://schema.org",{"og:url":83,"og:type":125,"og:title":65,"og:site_name":95,"og:description":67},"article",{"robots":127,"canonical":83},"index,follow",{"doc_id":129,"site_id":62},352696,1790377835,{"code":4,"msg":5,"data":132},{"doc_id":129,"user_id":133,"nickname":92,"user_avatar":134,"doc_module":4,"category_id":39,"category_name":40,"doc_title":65,"doc_description":67,"doc_content":135,"file_id":136,"file_url":137,"file_type":138,"file_size":139,"view_count":105,"is_deleted":4,"is_public":8,"is_downloadable":8,"audit_status":8,"page_count":140,"language":141,"language_code":63,"site_id":62,"html_lang":63,"table_of_contents":142,"faqs":143,"seo_title":144,"seo_description":67,"update_tm":145,"read_time":146,"preview_page_count":19,"preview_urls":147},962084928432,"https://ap-avatar.wpscdn.com/davatar_155a257f0dc6eb9ab79c44ca47cae57d","Author Manuscript Author Manuscript Author Manuscript Author Manuscript  \n\n| | HHS Public Access\u003Cbr>Author manuscript\u003Cbr>Comput Biol Med. Author manuscript; available in PMC 2026 April 07. |\n| --- | --- |\n\nPublished in final edited form as:  \nComput Biol Med. 2026 April 15; 206: 111603. doi:10.1016/j.compbiomed.2026.111603 .  \nEmpirical evaluation of variability and multi-institutional generalizability of deep learning survival models: application to renal cancer CT scans  \nBrennan Flannerya,* , Thomas DeSilvioa, Mohsen Haririb, Amir Reza Sadria, Nicholas Hellerc, Christopher Weightc, Satish E. Viswanatha,d  \naCase Western Reserve University, Department of Biomedical Engineering, 10900 Euclid Ave, Cleveland, OH 44106, USA  \nbCase Western Reserve University, Department of Computer and Data Sciences, 10900 Euclid Ave, Cleveland, OH 44106, USA  \ncCleveland Clinic, Department of Urology, 9500 Euclid Ave, Cleveland, OH 44106, USA dLouis Stokes Veteran Affairs Medical Center, 10701 East Blvd, Cleveland, OH 44106, USA  \nAbstract  \nDeep learning survival (DLS) modeling has shown significant promise across multiple disease conditions for predicting patient outcomes via biomedical data, but remain relatively underexplored in conjunction with medical imaging. This poses the question of what methodological choices promote model generalization, especially when predicting continuous survival outcomesin external validation cohorts. In this work, we systematically examined the impact of data partitioning, data order, model initialization, and augmentation strategies on the generalization of CT-based DLS survival models in renal cancers. A multi-institutional CT cohort was assembled, leveraging public repositories such as TCGA-KIRC, TCGA-KICH, TCGA-KIRP, and KITS19;  \ntotaling 525 patients from across 9 institutions. A 3D ResNet-18 model was trained to predict  \noverall patient survival with a Cox loss function and evaluated across three controlled experiments:(i) random versus covariate-balanced data partitioning,(ii) randomizing vs fixing data ordering and initial model weights, and (iii) varying intensity, spatial, and noise based augmentations;  \nevaluated in terms of concordance index (c-index) and hazard ratio (HR) . Intelligent covariatebalanced data partitioning demonstrated optimal performance on an external validation set (c-index 0.74, HR 5.08) compared to random partitioning (c-index 0.56, HR 0.29) . Different  \nThis is an open access article under the CC BY-NC license ([http://creativecommons.org/licenses/by-nc/4.0/](http://creativecommons.org/licenses/by-nc/4.0/)).  \n*[Corresponding author. bxf169@case.edu](Corresponding author. bxf169@case.edu) (B. Flannery), [satish.viswanath@emory.edu](satish.viswanath@emory.edu) (S.E. Viswanath) .  \nCRediT authorship contribution statement  \nBrennan Flannery: Writing – review & editing, Writing – original draft, Visualization, Validation, Software, Methodology, Investigation, Formal analysis, Data curation, Conceptualization. Thomas DeSilvio: Visualization, Conceptualization. Mohsen Hariri: Conceptualization. Amir Reza Sadri: Data curation. Nicholas Heller: Conceptualization. Christopher Weight:  \nConceptualization. Satish E. Viswanath: Writing – review & editing, Supervision, Resources, Project administration, Methodology, Funding acquisition, Conceptualization.  \nEthical statement  \nThis study was considered exempt from IRB review as it used public repositories of de-identified medical images only. No human subjects were directly involved, and no new data collection was performed.  \nAuthor Manuscript Author Manuscript Author Manuscript Author Manuscript  \nFlannery et al. Page 2  \nmodel initializations significantly impacted the average performance and variance across replicate trainings, while data ordering had little to no effect. Combining multiple augmentations produced the best performance on external validation (c-index +4.76%, HR +44.39%) compared to no augmentations, while most impac","cbCaimGJMiQ5whWr","https://ap.wps.com/l/cbCaimGJMiQ5whWr","pdf",2321548,28,"English","# Abstract\n# Keywords\n# 1. Introduction","[{\"question\":\"What problem does the study address regarding deep learning survival (DLS) models?\",\"answer\":\"It examines which methodological choices improve model generalization for CT-based DLS, particularly for continuous survival outcome prediction in external validation cohorts.\"},{\"question\":\"Which factors did the study systematically evaluate for generalization?\",\"answer\":\"Data partitioning, data order, model initialization, and image augmentation strategies were tested using controlled experiments.\"},{\"question\":\"What were the key findings on external validation performance?\",\"answer\":\"Covariate-balanced partitioning outperformed random partitioning, model initialization significantly affected average performance and variance, and combining augmentations improved external validation, with additive Gaussian noise being the most impactful individual augmentation.\"}]","Empirical evaluation of variability and multi-institutional generalizability of deep learning survival models - application to renal cancer CT scans | PDF",1790100938,71,[148,152,156,160],{"page_no":8,"url":149,"width":150,"height":151},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-16/5edca51d0ba041997a9fee3d71202398.webp",983,1272,{"page_no":14,"url":153,"width":154,"height":155},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-16/7e7b4159843d9611ec93c16f02478071.webp",982,1270,{"page_no":81,"url":157,"width":158,"height":159},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-16/d776e84aaae18bd5ec27e6155316014a.webp",981,1269,{"page_no":19,"url":161,"width":162,"height":163},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-16/f9364db84cea8a7253d9b2c99b09483f.webp",980,1268]