[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"detail-sidebar-cat-0-en-105":3,"doc-seo-384237-105":59,"doc-detail-384237-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","saliency-of-breast-lesions-in-breast-cancer-detection-using-artificial-intelligence","Saliency of breast lesions in breast cancer detection using artificial intelligence","","Saliency maps are used to examine how breast lesions influence the decision-making of deep learning AI systems for breast cancer detection on screening mammograms. Mammograms from 191 screen-detected breast cancer patients and 191 age- and system-matched healthy controls were collected. Two radiologists manually segmented lesions on CC and MLO views. Detection performance of four AI systems was evaluated using AUC with 95% confidence intervals, and saliency-derived areas of interest were compared to lesions using Dice similarity coefficient.",{"@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":35,"@type":76,"position":81},"https://docshare.wps.com/document/healthcare/",3,{"item":83,"name":65,"@type":76,"position":19},"https://docshare.wps.com/document/saliency-of-breast-lesions-in-breast-cancer-detection-using-artificial-intelligence/384237/",{"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/saliency-of-breast-lesions-in-breast-cancer-detection-using-artificial-intelligence/384237.png","ImageObject",300,407,{"name":92,"@type":93},"Kurz","Person",{"url":74,"name":95,"@type":96},"DocShare","Organization","application/pdf","2026-10-27","2026-09-24",true,{"@type":102,"interactionType":103,"userInteractionCount":39},"InteractionCounter",{"@type":104},"ViewAction",{"@type":106,"mainEntity":107},"FAQPage",[108,114,118],{"name":109,"@type":110,"acceptedAnswer":111},"How were the mammograms and controls selected for the study?","Question",{"text":112,"@type":113},"Mammograms were retrospectively collected from 191 women with screen-detected breast cancer and 191 healthy controls matched by age and mammographic system.","Answer",{"name":115,"@type":110,"acceptedAnswer":116},"What was the role of radiologists in the analysis?",{"text":117,"@type":113},"Two radiologists manually segmented the breast lesions in CC and MLO views to provide reference lesion localization.",{"name":119,"@type":110,"acceptedAnswer":120},"How did the researchers evaluate AI performance and saliency overlap with lesions?",{"text":121,"@type":113},"AI detection performance was measured using AUC with 95% confidence intervals, while the overlap between saliency-derived areas of interest and segmented lesions was assessed using Dice similarity coefficient (DSC).","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},384237,1790318458,{"code":4,"msg":5,"data":131},{"doc_id":128,"user_id":132,"nickname":92,"user_avatar":133,"doc_module":4,"category_id":34,"category_name":35,"doc_title":65,"doc_description":67,"doc_content":134,"file_id":135,"file_url":136,"file_type":137,"file_size":138,"view_count":39,"is_deleted":4,"is_public":8,"is_downloadable":8,"audit_status":8,"page_count":44,"language":139,"language_code":63,"site_id":62,"html_lang":63,"table_of_contents":140,"faqs":141,"seo_title":142,"seo_description":67,"update_tm":143,"read_time":144},2336478945635,"https://ap-avatar.wpscdn.com/davatar_6f874abed73319feea01a86fa6f0fab8","[www. nature.com/scientificreports](www. nature.com/scientificreports)  \nOPEN  \nSaliency of breast lesions in breast cancer detection using artificial intelligence  \nSaid Pertuz1, David Ortega1, Érika Suarez1, William Cancino1, Gerson Africano1, Irina Rinta‑Kiikka2,3, OtsoArponen2,3*, Sara Paris4 & Alfonso Lozano4  \nThe analysis of mammograms using artificial intelligence (AI) has shown great potential for assisting breast cancer screening. We use saliency maps to study the role of breast lesions in the decision‑making process of AI systems for breast cancer detection in screening mammograms. We retrospectively collected mammograms from 191 women with screen‑detected breast cancer and 191 healthy controls matched by age and mammographic system. Two radiologists manually segmented the breast lesions in the mammograms from CC and MLO views. We estimated the detection performance of four deep learning‑based AI systems using the area under the ROC curve (AUC) with a 95% confidence interval (CI). We used automatic thresholding on saliency maps from the AI systems to identify the areas of interest on the mammograms. Finally, we measured the overlap between these areas of interest and the segmented breast lesions using Dice’s similarity coefficient (DSC). The detection performance of the AI systems ranged from low to moderate (AUCs from 0.525 to 0.694) . The overlap between the areas of interest and the breast lesions was low for all the studied methods (median DSC from 4.2% to 38.0%). The AI system with the highest cancer detection performance (AUC = 0.694, CI 0.662–0.726) showed the lowest overlap (DSC = 4.2%) with breast lesions. The areas of interest found by saliency analysis of the AI systems showed poor overlap with breast lesions.  \nThese results suggest that AI systems with the highest performance do not solely rely on localized breast lesions for their decision‑making in cancer detection; rather, they incorporate information from large image regions. This work contributes to the understanding of the role of breast lesions in cancer detection using AI.  \nIn recent years, artificial intelligence (AI) systems have shown great potential in the analysis of screening mammograms for breast cancer detection1. One of the main difficulties for their adoption in clinical practice is the limited understanding of how these systems make their decisions2. Specifically, AI systems often involve complex computational layers with millions of parameters, which hinders the interpretation of the system’s output by humans. This difficulty is often referred to as the “black box” problem of AI3,4.  \nTo overcome this problem, explainable AI (XAI) has been developed to provide methods for understanding the decision-making of AI systems. A recent survey found that 37% of diagnostic test studies in radiology incorporate some form of XAI5. The main approaches to XAI can be classified depending on their use of visualization, semantics, or counterexamples. Visualization methods, also known as visual explanation or saliency mapping methods, are the most common form of XAI6. Saliency mapping is used to generate “heatmaps” that highlight the areas in medical images that most influence the prediction of the AI system7. Semantic methods, also known as textual explanation methods, provide text descriptions. Counterexamples or example-based explanation methods provide examples related to the data under analysis. A comparison of these methods of explanation is beyond the scope of this work. For a detailed review of XAI methods, we refer the reader to (Reyes et al.3) and (Van der Velden et al.6) .  \nIn this work, we use saliency analysis to compare the overlap between areas of interest from AI systems against the localization of breast lesions segmented by expert radiologists in screening mammograms. In the analysis of mammograms, the presence of suspicious regions or lesions plays a pivotal role in the decision-making process of human readers8. In turn, because ","cbCaivaAJahy9Ua2","https://ap.wps.com/l/cbCaivaAJahy9Ua2","pdf",1408496,"English","# Materials and methods\n## Imaging data and lesion segmentation","[{\"question\":\"How were the mammograms and controls selected for the study?\",\"answer\":\"Mammograms were retrospectively collected from 191 women with screen-detected breast cancer and 191 healthy controls matched by age and mammographic system.\"},{\"question\":\"What was the role of radiologists in the analysis?\",\"answer\":\"Two radiologists manually segmented the breast lesions in CC and MLO views to provide reference lesion localization.\"},{\"question\":\"How did the researchers evaluate AI performance and saliency overlap with lesions?\",\"answer\":\"AI detection performance was measured using AUC with 95% confidence intervals, while the overlap between saliency-derived areas of interest and segmented lesions was assessed using Dice similarity coefficient (DSC).\"}]","Saliency of breast lesions in breast cancer detection using artificial intelligence | PDF",1790261800,23]