Artificial intelligence and imaging for diagnostic and treatment challenges in breast care : second Deep Breast Workshop, Deep-Breath 2025, held in conjunction with MICCAI 2025, Daejeon, South Korea, September 23, 2025, Proceedings. Dee...最新文献

筛选

英文中文

Longitudinal Mammogram Exam-Based Breast Cancer Diagnosis Models: Vulnerability to Adversarial Attacks. 基于纵向乳房x光检查的乳腺癌诊断模型：对抗性攻击的脆弱性。

Artificial intelligence and imaging for diagnostic and treatment challenges in breast care : second Deep Breast Workshop, Deep-Breath 2025, held in conjunction with MICCAI 2025, Daejeon, South Korea, September 23, 2025, Proceedings. Dee... Pub Date : 2026-01-01 Epub Date: 2025-09-21 DOI: 10.1007/978-3-032-05559-0_35

Zhengbo Zhou, Degan Hao, Dooman Arefan, Margarita Zuley, Jules Sumkin, Lars Grimm, James Joshi, Shandong Wu

{"title":"Longitudinal Mammogram Exam-Based Breast Cancer Diagnosis Models: Vulnerability to Adversarial Attacks.","authors":"Zhengbo Zhou, Degan Hao, Dooman Arefan, Margarita Zuley, Jules Sumkin, Lars Grimm, James Joshi, Shandong Wu","doi":"10.1007/978-3-032-05559-0_35","DOIUrl":"10.1007/978-3-032-05559-0_35","url":null,"abstract":"<p><p>In breast cancer detection and diagnosis, the longitudinal analysis of mammogram images is crucial. Contemporary models excel in detecting temporal imaging feature changes, thus enhancing the learning process over sequential imaging exams. Yet, the resilience of these longitudinal models against adversarial attacks remains underexplored. In this paper, we propose a novel blackbox attack approach that capitalizes on the feature-level relationship between two sequential mammogram exams of a longitudinal model, guided by both cross-entropy loss and distance metric learning, to achieve significant attack efficacy as implemented using attack transferring in a black-box attacking manner. We perform experiments on a cohort of 590 breast cancer patients (each has two sequential mammogram exams) in a case-control setting. Results show that our proposed method surpasses several state-of-the-art adversarial attacks in fooling the diagnosis models to give opposite outputs. Our method remains effective even if the model is trained by employing existing defense mechanisms for adversarial attacks.</p>","PeriodicalId":524259,"journal":{"name":"Artificial intelligence and imaging for diagnostic and treatment challenges in breast care : second Deep Breast Workshop, Deep-Breath 2025, held in conjunction with MICCAI 2025, Daejeon, South Korea, September 23, 2025, Proceedings. Dee...","volume":"16142 ","pages":"351-361"},"PeriodicalIF":0.0,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13678561/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149229846","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0