[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-85247-en":3,"doc-seo-85247-105":30,"detail-sidebar-cat-0-en-105":91},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":20,"is_deleted":4,"is_public":21,"is_downloadable":21,"audit_status":21,"page_count":22,"language":23,"language_code":24,"site_id":25,"html_lang":24,"table_of_contents":26,"faqs":27,"seo_title":13,"seo_description":14,"update_tm":28,"read_time":29},85247,1374391974564,"Clementine","https://ap-avatar.wpscdn.com/avatar/14000253aa45c000a9e?x-image-process=image/resize,m_fixed,w_180,h_180&k=1779874745381141002",8,"Research & Report","Toward Efficient Weakly Supervised Semantic Segmentation Using Only Low-Magnification Histopathological Images","Whole-slide images (WSIs) deliver rich tissue and cellular detail, yet high-magnification storage and transmission are resource-intensive, and pixel-level annotation is costly. This work studies weakly supervised histopathological segmentation using low-magnification inputs with limited labels. A controlled benchmark simulates resolution degradation from high-resolution patches, then reconstructs them to the original size via interpolation and deep learning reconstruction. Results show reconstruction metrics alone cannot reliably forecast segmentation quality, revealing a critical degradation threshold for small-structure localization.","Toward Efficient Weakly Supervised Semantic Segmentation Using Only Low-Magnification Histopathological Images  \nDung Minh Do 1,2,* , Nhat-Thanh Huynh 1,2,* , Duc Minh Huynh 1,2 , Doanh C. Bui 1,2 , and Khang Nguyen 1,2,†  \n1 University of Information Technology, Ho Chi Minh City, Vietnam  \n2 Vietnam National University Ho Chi Minh City, Ho Chi Minh City, Vietnam  \n{23520326, [23521440](23521440}@gm.uit.edu.vn)[}](23521440}@gm.uit.edu.vn)[@gm.uit.edu.vn](23521440}@gm.uit.edu.vn), {duchm, doanhbc, [khangnttm](khangnttm}@uit.edu.vn)[}](khangnttm}@uit.edu.vn)[@uit.edu.vn](khangnttm}@uit.edu.vn)  \n*These authors contributed equally to this work.  \n†Corresponding author: [khangnttm@uit.edu.vn](khangnttm@uit.edu.vn).  \narXiv :2607 . 10783v1 [ cs .CV] 12 Jul 2026  \nAbstract—Whole-slide images (WSIs) provide rich tissue-level and cellular-level information, but storing and transmitting highmagnification pathology data is resource-intensive. Moreover, annotating WSIs at the pixel level is labor-intensive and timeconsuming. Therefore, it is important to investigate whether lowmagnification pathology images with limited annotations (i.e., image-level instead of pixel-level labels) can achieve performance comparable to high-magnification images. This paper presents a systematic benchmark study on weakly supervised histopathological image segmentation under different low-resolution storage settings. Starting from high-resolution image patches, we simulate lower-magnification inputs and reconstruct them to the original size using interpolation and deep learning-based reconstruction methods before applying the weakly-supervised segmentation pipeline. This framework enables a quantitative evaluation of how weakly supervised methods respond to different levels of resolution degradation. Experimental results show that reconstruction quality metrics alone are insufficient to predict downstream segmentation performance. In particular, the study identifies a critical degradation point where the localization of small-scale structures declines significantly. These findings provide practical guidance for designing efficient digital pathology storage systems while maintaining reliable automated analysis. Code is available at [https://github.com/Dung-Dx/LowMagWSS](https://github.com/Dung-Dx/LowMagWSS)  \nIndex Terms—Weakly supervised segmentation, histopathology images, super-resolution, low-resolution storage.  \nI. INTRODUCTION  \nThe rapid adoption of digital pathology has led to a substantial increase in whole-slide image (WSI) data. Highmagnification pathology images preserve detailed tissue morphology, cellular structures, and boundary information that are important for computational pathology tasks such as tissue classification and semantic segmentation [1] . However, storing and transferring high-resolution pathology images requires considerable storage capacity and bandwidth. This creates a practical need for storage-efficient computational pathology, where images may be stored at lower magnification while still supporting reliable downstream analysis.  \nWeakly supervised semantic segmentation (WSS) is an attractive setting for computational pathology because it reduces the need for dense pixel-level annotations. Instead of requiring detailed masks, WSS methods can learn from weaker labels  \nsuch as image-level or patch-level classification labels [2], [3] . However, WSS often depends on indirect localization cues, such as class activation maps or pseudo masks, which maybe sensitive to image degradation [4], [5] . When pathology images are stored at lower resolution, fine tissue boundaries and local textures may be lost, potentially reducing the quality of weak supervision and the final segmentation results.  \nSuper-resolution (SR) provides a possible way to reconstruct high-resolution pathology images from low-resolution inputs. Simple interpolation methods are computationally cheap but cannot recover lost high-frequency details. Learned patholo","cbCaidsJHV2P0mTc","https://ap.wps.com/l/cbCaidsJHV2P0mTc","pdf",7418518,2,1,6,"English","en",105,"# Introduction\n# Related Work\n## Weakly Supervised Segmentation in Histopathology","[{\"question\":\"Why is weakly supervised semantic segmentation important for histopathology?\",\"answer\":\"It reduces the need for dense pixel-level masks by learning from weaker labels such as image-level or patch-level supervision, which are cheaper to obtain.\"},{\"question\":\"How does the paper simulate low-magnification storage for evaluation?\",\"answer\":\"It starts from high-resolution image patches, downsamples them to mimic lower-magnification storage, reconstructs them back to the original size using interpolation or learned reconstruction, and then applies a fixed weakly supervised segmentation pipeline.\"},{\"question\":\"What is the main limitation of relying only on reconstruction quality metrics?\",\"answer\":\"Reconstruction quality metrics alone cannot predict downstream segmentation performance; small-scale structure localization degrades significantly beyond a critical degradation point.\"}]",1784202056,15,{"code":4,"msg":31,"data":32},"ok",{"site_id":25,"language":24,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":28},"toward-efficient-weakly-supervised-semantic-segmentation-using-only-low-magnification-histopathological-images","",{"@graph":36,"@context":85},[37,53,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,47,50],{"item":41,"name":42,"@type":43,"position":21},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":20},"https://docshare.wps.com/document/","Document",{"item":48,"name":12,"@type":43,"position":49},"https://docshare.wps.com/document/research-report/",3,{"item":51,"name":13,"@type":43,"position":52},"https://docshare.wps.com/document/toward-efficient-weakly-supervised-semantic-segmentation-using-only-low-magnification-histopathological-images/85247/",4,{"url":51,"name":13,"@type":54,"author":55,"headline":13,"publisher":57,"fileFormat":60,"inLanguage":24,"description":14,"dateModified":61,"datePublished":62,"encodingFormat":60,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":56},"Person",{"url":41,"name":58,"@type":59},"DocShare","Organization","application/pdf","2026-07-24","2026-07-16",true,{"@type":65,"interactionType":66,"userInteractionCount":20},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"Why is weakly supervised semantic segmentation important for histopathology?","Question",{"text":75,"@type":76},"It reduces the need for dense pixel-level masks by learning from weaker labels such as image-level or patch-level supervision, which are cheaper to obtain.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How does the paper simulate low-magnification storage for evaluation?",{"text":80,"@type":76},"It starts from high-resolution image patches, downsamples them to mimic lower-magnification storage, reconstructs them back to the original size using interpolation or learned reconstruction, and then applies a fixed weakly supervised segmentation pipeline.",{"name":82,"@type":73,"acceptedAnswer":83},"What is the main limitation of relying only on reconstruction quality metrics?",{"text":84,"@type":76},"Reconstruction quality metrics alone cannot predict downstream segmentation performance; small-scale structure localization degrades significantly beyond a critical degradation point.","https://schema.org",{"og:url":51,"og:type":87,"og:title":13,"og:site_name":58,"og:description":14},"article",{"robots":89,"canonical":51},"index,follow",{"doc_id":7,"site_id":25},{"code":4,"msg":5,"data":92},[93,97,101,105,110,114,119,122,127,130,134],{"id":21,"doc_module":4,"doc_module_name":46,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":20,"doc_module":4,"doc_module_name":46,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":52,"doc_module":4,"doc_module_name":46,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":46,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":22,"doc_module":4,"doc_module_name":46,"category_name":111,"show_sort_weight":112,"slug":113},"Technology",50,"technology",{"id":115,"doc_module":4,"doc_module_name":46,"category_name":116,"show_sort_weight":117,"slug":118},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":120,"slug":121},30,"research-report",{"id":123,"doc_module":4,"doc_module_name":46,"category_name":124,"show_sort_weight":125,"slug":126},9,"Religion & Spirituality",20,"religion-spirituality",{"id":125,"doc_module":4,"doc_module_name":46,"category_name":128,"show_sort_weight":125,"slug":129},"World Cup","world-cup",{"id":131,"doc_module":4,"doc_module_name":46,"category_name":132,"show_sort_weight":131,"slug":133},10,"Lifestyle","lifestyle",{"id":135,"doc_module":4,"doc_module_name":46,"category_name":136,"show_sort_weight":106,"slug":137},19,"General","general"]