[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-seo-382282-105":3,"detail-sidebar-cat-0-en-105":81,"doc-detail-382282-en":130},{"code":4,"msg":5,"data":6},0,"ok",{"site_id":7,"language":8,"slug":9,"title":10,"keywords":11,"description":12,"schema_data":13,"social_meta":74,"head_meta":76,"extra_data":78,"updated_unix":80},105,"en","automatic-classification-of-histopathology-images-across-multiple-cancers-based-on-heterogeneous-transfer-learning","Automatic Classification of Histopathology Images across Multiple Cancers Based on Heterogeneous Transfer Learning","","Heterogeneous transfer learning is evaluated to overcome histopathology AI’s dependence on large, heavily labeled datasets for each cancer type. A source domain model is trained on 28,634 colorectal patches, while two target domain models use 1,000 sentinel lymph node patches and 1,008 breast patches. Domain adaptation reduces feature distribution gaps and employs maximum mean discrepancy for knowledge transfer. Results show HTL improves AUC over supervised learning on sentinel lymph node and breast tasks.",{"@graph":14,"@context":73},[15,34,56],{"@type":16,"itemListElement":17},"BreadcrumbList",[18,23,27,31],{"item":19,"name":20,"@type":21,"position":22},"https://docshare.wps.com","Home","ListItem",1,{"item":24,"name":25,"@type":21,"position":26},"https://docshare.wps.com/document/","Document",2,{"item":28,"name":29,"@type":21,"position":30},"https://docshare.wps.com/document/healthcare/","Healthcare",3,{"item":32,"name":10,"@type":21,"position":33},"https://docshare.wps.com/document/automatic-classification-of-histopathology-images-across-multiple-cancers-based-on-heterogeneous-transfer-learning/382282/",4,{"url":32,"name":10,"@type":35,"image":36,"author":41,"headline":10,"publisher":44,"fileFormat":47,"inLanguage":8,"description":12,"dateModified":48,"datePublished":49,"encodingFormat":47,"isAccessibleForFree":50,"interactionStatistic":51},"DigitalDocument",{"url":37,"@type":38,"width":39,"height":40},"https://docshare.wps.com/thumbnails/automatic-classification-of-histopathology-images-across-multiple-cancers-based-on-heterogeneous-transfer-learning/382282.png","ImageObject",300,407,{"name":42,"@type":43},"Eliana","Person",{"url":19,"name":45,"@type":46},"DocShare","Organization","application/pdf","2026-11-03","2026-09-24",true,{"@type":52,"interactionType":53,"userInteractionCount":55},"InteractionCounter",{"@type":54},"ViewAction",8,{"@type":57,"mainEntity":58},"FAQPage",[59,65,69],{"name":60,"@type":61,"acceptedAnswer":62},"What problem does the paper target in histopathology AI?","Question",{"text":63,"@type":64},"Current AI methods often specialize in a single cancer task, requiring extensive image collection and labeling for each cancer type, which creates a major data bottleneck.","Answer",{"name":66,"@type":61,"acceptedAnswer":67},"How is heterogeneous transfer learning used in the study?",{"text":68,"@type":64},"A source domain model is trained on colorectal patches, then two target domain models are trained using sentinel lymph node and breast patches. Heterogeneous domain adaptation reduces feature distribution differences and maximum mean discrepancy is used for knowledge transfer.",{"name":70,"@type":61,"acceptedAnswer":71},"What performance differences does the study report?",{"text":72,"@type":64},"HTL on 1,000 sentinel lymph node patches achieves higher average AUC than supervised learning on the same target data, with no significant difference in the comparable larger supervised setting. Similar improvements are reported for breast cancer.","https://schema.org",{"og:url":32,"og:type":75,"og:title":10,"og:site_name":45,"og:description":12},"article",{"robots":77,"canonical":32},"index,follow",{"doc_id":79,"site_id":7},382282,1790794813,{"code":4,"msg":82,"data":83},"success",[84,88,92,96,101,106,110,114,119,122,126],{"id":22,"doc_module":4,"doc_module_name":25,"category_name":85,"show_sort_weight":86,"slug":87},"Story & Novel",90,"story-novel",{"id":26,"doc_module":4,"doc_module_name":25,"category_name":89,"show_sort_weight":90,"slug":91},"Literature",80,"literature",{"id":33,"doc_module":4,"doc_module_name":25,"category_name":93,"show_sort_weight":94,"slug":95},"Exam",70,"exam",{"id":97,"doc_module":4,"doc_module_name":25,"category_name":98,"show_sort_weight":99,"slug":100},5,"Comic",60,"comic",{"id":102,"doc_module":4,"doc_module_name":25,"category_name":103,"show_sort_weight":104,"slug":105},6,"Technology",50,"technology",{"id":107,"doc_module":4,"doc_module_name":25,"category_name":29,"show_sort_weight":108,"slug":109},7,40,"healthcare",{"id":55,"doc_module":4,"doc_module_name":25,"category_name":111,"show_sort_weight":112,"slug":113},"Research & Report",30,"research-report",{"id":115,"doc_module":4,"doc_module_name":25,"category_name":116,"show_sort_weight":117,"slug":118},9,"Religion & Spirituality",20,"religion-spirituality",{"id":117,"doc_module":4,"doc_module_name":25,"category_name":120,"show_sort_weight":117,"slug":121},"World Cup","world-cup",{"id":123,"doc_module":4,"doc_module_name":25,"category_name":124,"show_sort_weight":123,"slug":125},10,"Lifestyle","lifestyle",{"id":127,"doc_module":4,"doc_module_name":25,"category_name":128,"show_sort_weight":97,"slug":129},19,"General","general",{"code":4,"msg":82,"data":131},{"doc_id":79,"user_id":132,"nickname":42,"user_avatar":133,"doc_module":4,"category_id":107,"category_name":29,"doc_title":10,"doc_description":12,"doc_content":134,"file_id":135,"file_url":136,"file_type":137,"file_size":138,"view_count":55,"is_deleted":4,"is_public":22,"is_downloadable":22,"audit_status":22,"page_count":139,"language":140,"language_code":8,"site_id":7,"html_lang":8,"table_of_contents":141,"faqs":142,"seo_title":143,"seo_description":12,"update_tm":144,"read_time":145},4398048949847,"https://ap-avatar.wpscdn.com/avatar/400002536579ef2da7f?_k=1778318612642679267","diagnostics  \nArticle  \nAutomatic Classification of Histopathology Images across Multiple Cancers Based on Heterogeneous Transfer Learning  \nKai Sun 1, Yushi Chen 2, Bingqian Bai 1, Yanhua Gao 3, Jiaying Xiao 1 and Gang Yu 1, *  \nCitation: Sun, K.; Chen, Y.; Bai, B.; Gao, Y.; Xiao, J.; Yu, G. Automatic Classification of Histopathology Images across Multiple Cancers Based on Heterogeneous Transfer Learning. Diagnostics 2023, 13, 1277 . [https://doi.org/10.3390/](https://doi.org/10.3390/)[ ](https://doi.org/10.3390/)diagnostics13071277  \nAcademic Editor: Sol±ne-Florence Kammerer-Jacquet  \nReceived: 10 February 2023  \nRevised: 7 March 2023  \nAccepted: 23 March 2023  \nPublished: 28 March 2023  \nCopyright: © 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license ([https://](https://)[ ](https://)[creativecommons.org/licenses/by/](creativecommons.org/licenses/by/)[ ](creativecommons.org/licenses/by/)[4.0/](4.0/)) .  \n1 Department of Biomedical Engineering, School of Basic Medical Sciences, Central South University, Changsha 410013, China  \n2 Department of Pathology, School of Basic Medical Sciences, Central South University, Changsha 410013, China  \n3 Department of Ultrasound, Shaanxi Provincial People's Hospital, Xi'an 710068, China  \n* [Correspondence: yugang@mail.csu.edu.cn](Correspondence: yugang@mail.csu.edu.cn)  \nAbstract: Background: Current artificial intelligence (AI) in histopathology typically specializeson a single task, resulting in a heavy workload of collecting and labeling a sufficient number of images for each type of cancer. Heterogeneous transfer learning (HTL) is expected to alleviate the data bottlenecks and establish models with performance comparable to supervised learning (SL) . Methods: An accurate source domain model was trained using 28,634 colorectal patches. Additionally, 1000 sentinel lymph node patches and 1008 breast patches were used to train two target domain models. The feature distribution difference between sentinel lymph node metastasis or breast cancer and CRC was reduced by heterogeneous domain adaptation, and the maximum mean difference between subdomains was used for knowledge transfer to achieve accurate classification across multiple cancers. Result: HTL on 1000 sentinel lymph node patches (L-HTL-1000) outperforms SL on 1000 sentinel lymph node patches (L-SL-1-1000) (average area under the curve (AUC) and standard deviation of L-HTL-1000 vs. L-SL-1-1000: 0.949 􀀆 0.004 vs. 0.931 􀀆 0.008, p value = 0.008) . There is no significant difference between L-HTL-1000 and SL on 7104 patches (L-SL-2-7104) (0.949 􀀆 0.004 vs. 0.948 􀀆 0.008, p value = 0.742) . Similar results are observed for breast cancer. B-HTL-1008 vs. B-SL-1-1008: 0.962 􀀆 0.017 vs. 0.943 􀀆 0.018, p value = 0.008; B-HTL-1008 vs. B-SL-2-5232: 0.962 􀀆 0.017 vs. 0.951 􀀆 0.023, p value = 0.148 . Conclusions: HTL is capable of building accurate AI models for similar cancers using a small amount of data based on a large dataset for a certain type of cancer. HTL holds great promise for accelerating the development of AI in histopathology.  \nKeywords: artificial intelligence in histopathology; heterogeneous transfer learning; cancer diagnosis; small datasets  \n1. Introduction  \nCancer is a leading cause of death worldwide, with common types including colorectal cancer (CRC), breast cancer, and others. In 2020, the global fatality rates for CRC and breast cancer were 9.4% and 6.9%, respectively [1] . Histopathology is an accurate method for diagnosing cancer [2], but it requires specialized knowledge and clinical experience from pathologists. Unfortunately, there is a shortage of pathologists worldwide, with the number of active pathologists decreasing by 17.53% in the United States from 2007 to 2017 [3] . In low-income countries, such as those in sub-Saharan Africa, there are fewer than one pathologist per 500,000 p","cbCaiaO4jLEKSaLV","https://ap.wps.com/l/cbCaiaO4jLEKSaLV","pdf",32504865,13,"English","# Introduction\n## Background on cancer burden and pathology challenges\n## Role of AI in histopathology\n## Limitations of current deep learning pipelines\n# Methods\n## Training source and target domain models\n## Heterogeneous transfer learning and knowledge transfer\n# Results\n## Performance on sentinel lymph node metastasis\n## Performance on breast cancer\n# Conclusions","[{\"question\":\"What problem does the paper target in histopathology AI?\",\"answer\":\"Current AI methods often specialize in a single cancer task, requiring extensive image collection and labeling for each cancer type, which creates a major data bottleneck.\"},{\"question\":\"How is heterogeneous transfer learning used in the study?\",\"answer\":\"A source domain model is trained on colorectal patches, then two target domain models are trained using sentinel lymph node and breast patches. Heterogeneous domain adaptation reduces feature distribution differences and maximum mean discrepancy is used for knowledge transfer.\"},{\"question\":\"What performance differences does the study report?\",\"answer\":\"HTL on 1,000 sentinel lymph node patches achieves higher average AUC than supervised learning on the same target data, with no significant difference in the comparable larger supervised setting. Similar improvements are reported for breast cancer.\"}]","Automatic Classification of Histopathology Images across Multiple Cancers Based on Heterogeneous Transfer Learning | PDF",1790249852,33]