arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2602.13930cs.CVcs.LG
收藏

MamaDino:一种用于乳腺癌3年风险预测的混合视觉模型

MamaDino: A Hybrid Vision Model for Breast Cancer 3-Year Risk Prediction

Fondazione Human Technopole(人类技术极地基金会)

Ruggiero Santeramo, Igor Zubarev, Florian Jug

AI总结:

MamaDino通过结合卷积和Transformer的归纳偏差及显式对侧不对称建模,在低分辨率乳腺X线摄影上实现了与Mirai相当的3年乳腺癌风险预测性能。

摘要:

乳腺癌筛查计划日益寻求从一刀切的间隔策略转向风险适应和个性化的策略。深度学习(DL)已经使基于图像的风险模型在1至5年预测方面优于传统临床模型,但领先系统(例如Mirai)通常使用卷积骨干网络、非常高的分辨率输入(>100万像素)和简单的多视图融合,且对对侧不对称性建模有限。我们假设结合互补的归纳偏差(卷积和基于Transformer的)与显式的对侧不对称性建模,即使在显著降低分辨率的乳腺X线摄影图像上,也能实现与最新3年风险预测性能相匹配,表明在更结构化的方式下使用更详细的图像可以恢复最新准确度。我们提出了MamaDino,一种针对乳腺X线摄影的多视图注意DINO模型。MamaDino融合了冻结的自监督DINOv3 ViT-S特征和在512x512分辨率上可训练的CNN编码器,并通过BilateralMixer聚合双侧乳腺信息以输出3年乳腺癌风险评分。我们在英国OPTIMAM(UK)的53,883名女性上进行训练,并在匹配的3年病例对照队列上进行评估:一个内部分布的测试集来自四个筛查地点,一个外部非分布队列来自未见过的地点。在乳腺层面,MamaDino在内部和外部测试中均与Mirai匹配,同时使用约13倍更少的输入像素。添加BilateralMixer提高了区分能力,内部测试的AUC达到0.736(vs 0.713),外部测试的AUC达到0.677(vs 0.666),在年龄、种族、扫描仪、肿瘤类型和分级方面表现一致。这些发现表明,显式的对侧建模和互补的归纳偏差使预测能够与Mirai相匹配,尽管在显著降低分辨率的乳腺X线摄影图像上进行操作。

英文摘要:

Breast cancer screening programmes increasingly seek to move from one-size-fits-all intervals to risk-adapted and personalized strategies. Deep learning (DL) has enabled image-based risk models with stronger 1- to 5-year prediction than clinical models, but leading systems (e.g., Mirai) typically use convolutional backbones, high-resolution inputs (>1M pixels) and simple multi-view fusion, with limited explicit modelling of contralateral asymmetry. We hypothesised that combining complementary inductive biases (convolutional and transformer-based) with explicit contralateral asymmetry modelling would allow us to match state-of-the-art 3-year risk prediction performance on lower-resolution mammograms, suggesting that using less detail in a more structured way can recover state-of-the-art accuracy. We present MamaDino, a mammography-aware multi-view attentional DINO model. MamaDino fuses frozen self-supervised DINOv3 ViT-S features with a trainable CNN encoder at 512x512 resolution and aggregates bilateral breast information via a BilateralMixer to output a 3-year breast cancer risk score. We train on 53,883 women from OPTIMAM (UK) and evaluate on matched 3-year case-control cohorts: an in-distribution test set from four screening sites and an external out-of-distribution cohort from an unseen site. At breast level, MamaDino achieves discrimination comparable to Mirai on internal and external tests while using ~13x fewer input pixels. Adding the BilateralMixer improves patient-level discrimination to AUC 0.736 (vs Mirai 0.713) in-distribution and 0.677 (vs Mirai 0.676) out-of-distribution, with overlapping confidence intervals. Performance was evaluated across age, ethnicity, scanner manufacturer, tumour type and grade. These findings suggest that complementary inductive biases and explicit contralateral modelling can achieve Mirai-comparable discrimination at lower resolution.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑