[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"detail-sidebar-cat-0-en-105":3,"doc-seo-360258-105":59,"doc-detail-360258-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","bidirectional-perceptual-multimodal-interaction-network-based-on-contrastive-learning-for-breast-cancer-pcr-prediction","Bidirectional Perceptual Multimodal Interaction Network Based on Contrastive Learning for Breast Cancer pCR Prediction","","Early and accurate prediction of pathological complete response (pCR) after neoadjuvant chemotherapy (NAC) is essential for personalized breast cancer treatment. Existing deep learning approaches are limited by tumor heterogeneity and semantic misalignment between high-dimensional dynamic contrast-enhanced MRI (DCE-MRI) and low-dimensional clinical data. A bidirectional contrastive multimodal network, BPMINet, fuses pre-NAC DCE-MRI with clinical information via bidirectional cross-modal attention, multimodal contrast-aware feature enhancement, and a dual-loss optimization. Across seven metrics on two multicenter datasets, BPMINet outperforms baselines and shows strong generalization.",{"@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":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/bidirectional-perceptual-multimodal-interaction-network-based-on-contrastive-learning-for-breast-cancer-pcr-prediction/360258/",{"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/bidirectional-perceptual-multimodal-interaction-network-based-on-contrastive-learning-for-breast-cancer-pcr-prediction/360258.png","ImageObject",300,407,{"name":92,"@type":93},"Adam","Person",{"url":74,"name":95,"@type":96},"DocShare","Organization","application/pdf","2026-11-13","2026-09-23",true,{"@type":102,"interactionType":103,"userInteractionCount":52},"InteractionCounter",{"@type":104},"ViewAction",{"@type":106,"mainEntity":107},"FAQPage",[108,114,118],{"name":109,"@type":110,"acceptedAnswer":111},"Why is predicting pathological complete response (pCR) after neoadjuvant chemotherapy important?","Question",{"text":112,"@type":113},"pCR is a key indicator of NAC efficacy and is linked to improved prognosis and greater breast-conserving potential. Early prediction supports timely personalized treatment decisions.","Answer",{"name":115,"@type":110,"acceptedAnswer":116},"What problem limits existing deep learning methods for pCR prediction?",{"text":117,"@type":113},"Tumor heterogeneity and semantic misalignment between high-dimensional DCE-MRI and low-dimensional clinical data reduce predictive performance and generalization.",{"name":119,"@type":110,"acceptedAnswer":120},"How does BPMINet improve multimodal integration for pCR prediction?",{"text":121,"@type":113},"BPMINet uses bidirectional cross-modal attention for semantic alignment, a multimodal contrast-aware feature enhancement module to increase discriminative power, and a dual-loss strategy to jointly optimize feature representation and pCR prediction.","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},360258,1790189272,{"code":4,"msg":5,"data":131},{"doc_id":128,"user_id":132,"nickname":92,"user_avatar":133,"doc_module":4,"category_id":39,"category_name":40,"doc_title":65,"doc_description":67,"doc_content":134,"file_id":135,"file_url":136,"file_type":137,"file_size":138,"view_count":52,"is_deleted":4,"is_public":8,"is_downloadable":8,"audit_status":8,"page_count":139,"language":140,"language_code":63,"site_id":62,"html_lang":63,"table_of_contents":141,"faqs":142,"seo_title":143,"seo_description":67,"update_tm":144,"read_time":145,"preview_page_count":19,"preview_urls":146},1374404737137,"https://ap-avatar.wpscdn.com/davatar_155a257f0dc6eb9ab79c44ca47cae57d","Article  \nBidirectional Perceptual Multimodal Interaction Network Based on Contrastive Learning for Breast Cancer pCR Prediction  \nJingjing Feng, Zongli Jiang and Jinli Zhang *  \nAcademic Editor: Sungheon Gene Kim  \nReceived: 29 January 2026  \nRevised: 3 May 2026  \nAccepted: 13 May 2026  \nPublished: 19 May 2026  \nCopyright: © 2026 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.  \nCollege of Computer Science, Beijing University of Technology, Beijing 100124, China;  \n[fengjingjing@emails.bjut.edu.cn](fengjingjing@emails.bjut.edu.cn) (J.F.); [jiangzl@bjut.edu.cn](jiangzl@bjut.edu.cn) (Z.J.)  \n* [Correspondence: jlzhangcs@bjut.edu.cn](Correspondence: jlzhangcs@bjut.edu.cn)  \nSimple Summary  \nAccurate prediction of chemotherapy response is vital for personalized breast cancer treatment. Existing artificial intelligence technologies often struggle to effectively integrate complex medical imaging with clinical data, restricting the performance of pCR prediction. To tackle this limitation, we propose a multimodal network named BPMINet. The proposed model leverages advanced deep learning techniques to fuse magnetic resonance imaging and clinical information, boosting pCR prediction performance. These findings can assist radiologists in making more precise clinical decisions. In the future, integrating diverse data for long-term monitoring will help establish a more reliable intelligent decision support system for breast cancer treatment.  \nAbstract  \nBackground/Objectives: Early and accurate prediction of pathological complete response (pCR) after neoadjuvant chemotherapy (NAC) is vital for personalized breast cancer treatment. However, existing deep learning methods are hampered by tumor heterogeneity and semantic misalignment between high-dimensional dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) and low-dimensional clinical data, which limits pCR prediction performance and generalization. This study addresses these challenges via a novel multimodal network. Methods: We propose a Bidirectional Perceptual Multimodal Interaction Network (BPMINet) based on contrastive learning. BPMINet integrates pre-NACDCE-MRI and clinical information through three core components: (1) we propose a bidirectional cross-modal attention (BiCMA) fusion mechanism to resolve semantic misalignment and facilitate effective multimodal feature fusion; (2) we design a multimodal contrastaware feature enhancement (MCFE) module as a key component tightly integrated into the pCR-oriented contrastive learning framework, which serves to boost discriminative power for pCR prediction and improve generalization performance on hard-to-classify samples;  \n(3) we adopt a dual-loss strategy to enable the collaborative optimization of discriminative feature representation and pCR prediction performance. Results: On two publicly available multicenter datasets, BPMINet outperformed all comparative methods across seven evaluation metrics: specifically, it surpassed the top-performing baseline by 5.17% in AUCand 5.24% in accuracy on the MAMA-MIA dataset. More notably, it achieved substantially larger gains of 11.72% in AUC and 7.38% in accuracy on the ISPY1 dataset. Conclusions: BPMINet achieves optimal pCR prediction performance, confirming its superiority and strong generalization ability for multimodal breast cancer pCR prediction.  \nKeywords: breast cancer; pCR prediction; multimodal fusion; contrastive learning  \n1. Introduction  \nBreast cancer is the most common malignant tumor in women worldwide, posing a major threat to women’s health due to rising morbidity and mortality. Notably, breast cancer exhibits high heterogeneity, characterized by distinct tumor morphology, specific cellular characteristics, and highly variable treatment responses among different patients [1] . Neoadjuvant chemotherapy (NAC) is pivotal for loc","cbCaioLUm5ieJqQM","https://ap.wps.com/l/cbCaioLUm5ieJqQM","pdf",3819807,34,"English","# Introduction\n## Background and motivation\n## Limitations of existing methods\n## Study aim and proposed approach\n# Methods\n## BPMINet overview\n## Bidirectional cross-modal attention (BiCMA)\n## Multimodal contrast-aware feature enhancement (MCFE)\n## Dual-loss optimization\n# Results\n## Performance on multicenter datasets\n## Evaluation metrics and improvements\n# Conclusions\n## Overall findings and generalization ability","[{\"question\":\"Why is predicting pathological complete response (pCR) after neoadjuvant chemotherapy important?\",\"answer\":\"pCR is a key indicator of NAC efficacy and is linked to improved prognosis and greater breast-conserving potential. Early prediction supports timely personalized treatment decisions.\"},{\"question\":\"What problem limits existing deep learning methods for pCR prediction?\",\"answer\":\"Tumor heterogeneity and semantic misalignment between high-dimensional DCE-MRI and low-dimensional clinical data reduce predictive performance and generalization.\"},{\"question\":\"How does BPMINet improve multimodal integration for pCR prediction?\",\"answer\":\"BPMINet uses bidirectional cross-modal attention for semantic alignment, a multimodal contrast-aware feature enhancement module to increase discriminative power, and a dual-loss strategy to jointly optimize feature representation and pCR prediction.\"}]","Bidirectional Perceptual Multimodal Interaction Network Based on Contrastive Learning for Breast Cancer pCR Prediction | PDF",1790139322,86,[147,151,155,159],{"page_no":8,"url":148,"width":149,"height":150},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-16/431e19e106289860ca28219602742d9a.webp",983,1390,{"page_no":14,"url":152,"width":153,"height":154},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-16/6a6ae1011c83250418d39495eb69b14c.webp",982,1388,{"page_no":81,"url":156,"width":157,"height":158},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-16/a7fe812e02e2ee4b6bd655740501c6c8.webp",981,1387,{"page_no":19,"url":160,"width":161,"height":162},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-16/f868dc5b8bd10b105a331e50a40d2056.webp",980,1385]