[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"detail-sidebar-cat-0-en-105":3,"doc-seo-141409-105":59,"doc-detail-141409-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","physical-color-calibration-of-digital-pathology-scanners-for-robust-ai-assisted-cancer-diagnosis","Physical Color Calibration of Digital Pathology Scanners - for Robust AI Assisted Cancer Diagnosis","","Digital pathology AI performance is constrained by technical inconsistencies in producing whole slide images, including scanner hardware differences and post-processing variations that can degrade model accuracy and limit clinical deployment. Changes in imaging workflow may further compromise diagnoses and threaten patient safety. This study evaluated whether physical color calibration of scanners can standardize WSI appearance and improve robustness. A color calibration slide was used across four laboratories, and results were tested on 1,161 prostate cancer WSIs, showing improved AI calibration and Gleason grading performance.",{"@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/physical-color-calibration-of-digital-pathology-scanners-for-robust-ai-assisted-cancer-diagnosis/141409/",{"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/physical-color-calibration-of-digital-pathology-scanners-for-robust-ai-assisted-cancer-diagnosis/141409.png","ImageObject",300,407,{"name":92,"@type":93},"River Wang","Person",{"url":74,"name":95,"@type":96},"DocShare","Organization","application/pdf","2026-10-26","2026-08-25",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},"Why does AI performance degrade in digital pathology?","Question",{"text":113,"@type":114},"Technical inconsistencies during scanning—such as differences in light sources, sensors, and quantizers—as well as post-processing variations can change WSI appearance and reduce AI model performance.","Answer",{"name":116,"@type":111,"acceptedAnswer":117},"What problem does this study address with physical color calibration?",{"text":118,"@type":114},"It evaluates whether physical color calibration can standardize WSI appearance across scanners, enabling more consistent and robust AI performance without relying on impractical site-by-site fine-tuning.",{"name":120,"@type":111,"acceptedAnswer":121},"How was the calibration approach evaluated for cancer diagnosis?",{"text":122,"@type":114},"A color calibration slide was used in four different laboratories, and its impact was assessed by testing an AI system for prostate cancer diagnosis on 1,161 whole slide images, focusing on Gleason grading performance.","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},141409,1787655619,{"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":130,"read_time":26,"preview_page_count":4},1099514067438,"https://ap-avatar.wpscdn.com/avatar/100002539ee87300030?x-image-process=image/resize,m_fixed,w_180,h_180&k=1780474512215547542","Physical Color Calibration of Digital Pathology Scanners for Robust Artificial Intelligence Assisted Cancer Diagnosis  \nXiaoyi Ji1 , Richard Salmon2 , Nita Mulliqi1 , Umair Khan3 , Yinxi Wang1 , Anders Blilie4 ,5 , Henrik Olsson1 , Bodil Ginnerup Pedersen6 , Karina Dalsgaard Sørensen7 ,8 , Benedicte Parm Ulhøi9 , Svein  \nR. Kjosavik4 , 10 , Emilius A. M. Janssen4 ,5 , Mattias Rantalainen1 , Lars Egevad11 , Pekka Ruusuvuori3 , 12 , Martin Eklund1 , Kimmo Kartasalo1  \n1. Department of Medical Epidemiology and Biostatistics, Karolinska Institutet, Stockholm, Sweden  \n2. FFEI Life Science, FFEI Ltd, Hemel Hempstead, UK  \n3. Institute of Biomedicine, University of Turku, Turku, Finland  \n4. Department of Pathology, Stavanger University Hospital, Stavanger, Norway  \n5. Faculty of Science and Technology, University of Stavanger, Stavanger, Norway  \n6. Department of Radiology, Aarhus University Hospital, Aarhus, Denmark  \n7. Department of Molecular Medicine, Aarhus University Hospital, Aarhus, Denmark  \n8. Department of Clinical Medicine, Aarhus University, Aarhus, Denmark  \n9. Department of Pathology, Aarhus University Hospital, Aarhus, Denmark  \n10. Faculty of Health Sciences, University of Stavanger, Stavanger, Norway  \n11. Department of Oncology and Pathology, Karolinska Institutet, Stockholm, Sweden  \n12. Faculty of Medicine and Health Technology, Tampere University, Tampere, Finland  \nCorresponding author: Kimmo Kartasalo, [kimmo.kartasalo@ki.se](kimmo.kartasalo@ki.se).  \nThe potential of artificial intelligence (AI) in digital pathology is limited by technical inconsistencies in the production of whole slide images (WSIs) , leading to degraded AI performance and posing a challenge for widespread clinical application as fine-tuning algorithms for each new site is impractical. Changes in the imaging workflow can also lead to compromised diagnoses and patient safety risks. We evaluated whether physical color calibration of scanners can standardize WSI appearance and enable robust AI performance. We employed a color calibration slide in four different laboratories and evaluated its impact on the performance of an AI system for prostate cancer diagnosis on 1 , 161 WSIs. Color standardization resulted in consistently improved AI model calibration and significant improvements in Gleason grading performance. The study demonstrates that physical color calibration provides a potential solution to the variation introduced by different scanners, making AI-based cancer diagnostics more reliable and applicable in clinical settings.  \nIntroduction  \nApplication of artificial intelligence (AI) to digital pathology has achieved highly accurate performance on tasks such as cancer detection and grading1 ,2 , prognostication3 ,4 , and prediction of molecular biomarkers5. However, it has been noticed that technical inconsistency during the scanning process introduced by hardware variance in light sources, sensors and quantizers6–8 , can degrade AI model performance9 , 10. Variations in image appearance can also be caused by different post-processing operations applied by the scanning software to the raw image data. This is a considerable hurdle to widespread clinical application of AI models, since tweaking of the models at each new site where they are to be deployed is not feasible. Site-specific fine-tuning, which is commonplace for currently commercially available algorithms, is also problematic in terms of medical device regulations and validation of the algorithms’ real-world performance.  \nEven if an algorithm would have been fine-tuned to a given laboratory, unexpected behavior can arise due to changes in the imaging workflow, such as damage or wear of scanner components, or system updates of scanning software. In clinical use, this can lead to incorrect diagnoses and poses a threat to patient safety. Statistical methods, such as conformal prediction11 , have been proposed for detecting such changes in the data generation process, indicating","cbCaisLfMU7wsrRM","https://ap.wps.com/l/cbCaisLfMU7wsrRM","pdf",1500330,24,"English","# Introduction\n## Background: AI in digital pathology and sources of inconsistency\n## Challenge: site-specific fine-tuning and clinical workflow changes\n## Prior approaches: data augmentation and computational stain normalization\n## Study focus: physical color calibration for robust AI performance","[{\"question\":\"Why does AI performance degrade in digital pathology?\",\"answer\":\"Technical inconsistencies during scanning—such as differences in light sources, sensors, and quantizers—as well as post-processing variations can change WSI appearance and reduce AI model performance.\"},{\"question\":\"What problem does this study address with physical color calibration?\",\"answer\":\"It evaluates whether physical color calibration can standardize WSI appearance across scanners, enabling more consistent and robust AI performance without relying on impractical site-by-site fine-tuning.\"},{\"question\":\"How was the calibration approach evaluated for cancer diagnosis?\",\"answer\":\"A color calibration slide was used in four different laboratories, and its impact was assessed by testing an AI system for prostate cancer diagnosis on 1,161 whole slide images, focusing on Gleason grading performance.\"}]","Physical Color Calibration of Digital Pathology Scanners - for Robust AI Assisted Cancer Diagnosis | PDF"]