在低类间变异性下挑战 DINOv3 基础模型：胎儿脑部超声的案例研究
Challenging DINOv3 Foundation Model under Low Inter-Class Variability: A Case Study on Fetal Brain Ultrasound
AI总结：
本研究首次评估了基础模型在低类间变异性胎儿脑部超声成像中的表现，构建了多中心基准 FetalUS-188K，发现与自然图像预训练相比，域特定自监督预训练能显著提升 DINOv3 对高度重叠解剖切面的区分能力。
摘要：
目的：本研究首次全面评估了基础模型在低类间变异性条件下的胎儿超声（US）成像中的表现。尽管 DINOv3 等近期视觉基础模型在医学领域展现出卓越的迁移能力，但其区分解剖结构相似区域的能力尚未得到系统研究。我们通过聚焦胎儿脑部标准切面——丘脑切面（TT）、脑室切面（TV）和小脑切面（TC）——来填补这一空白，这些切面具有高度重叠的解剖特征，对可靠的生物测量评估构成了关键挑战。方法：为确保评估的公平性与可复现性，我们将所有公开的胎儿超声数据集整理并聚合为一个统一的多中心基准 FetalUS-188K，包含来自异构采集环境的超过 188,000 张标注图像。DINOv3 以自监督方式进行预训练，以学习超声感知表示。随后通过标准化适配协议评估所学特征，包括冻结骨干网络的线性探测和全量微调，并在两种初始化方案下进行：(i) 在 FetalUS-188K 上预训练；(ii) 从自然图像 DINOv3 权重初始化。结果：在胎儿超声数据上预训练的模型始终优于基于自然图像初始化的模型，加权 F1 分数提升最高达 20%。域自适应预训练使网络能够保留细微的回声和结构线索，这对于区分 TV 等中间切面至关重要。结论：结果表明，通用基础模型在低类间变异性下无法泛化，而域特定预训练对于在胎儿脑部超声成像中实现稳健且临床可靠的表示至关重要。
英文摘要：
Purpose: This study provides the first comprehensive evaluation of foundation models in fetal ultrasound (US) imaging under low inter-class variability conditions. While recent vision foundation models such as DINOv3 have shown remarkable transferability across medical domains, their ability to discriminate anatomically similar structures has not been systematically investigated. We address this gap by focusing on fetal brain standard planes--transthalamic (TT), transventricular (TV), and transcerebellar (TC)--which exhibit highly overlapping anatomical features and pose a critical challenge for reliable biometric assessment. Methods: To ensure a fair and reproducible evaluation, all publicly available fetal ultrasound datasets were curated and aggregated into a unified multicenter benchmark, FetalUS-188K, comprising more than 188,000 annotated images from heterogeneous acquisition settings. DINOv3 was pretrained in a self-supervised manner to learn ultrasound-aware representations. The learned features were then evaluated through standardized adaptation protocols, including linear probing with frozen backbone and full fine-tuning, under two initialization schemes: (i) pretraining on FetalUS-188K and (ii) initialization from natural-image DINOv3 weights. Results: Models pretrained on fetal ultrasound data consistently outperformed those initialized on natural images, with weighted F1-score improvements of up to 20 percent. Domain-adaptive pretraining enabled the network to preserve subtle echogenic and structural cues crucial for distinguishing intermediate planes such as TV. Conclusion: Results demonstrate that generic foundation models fail to generalize under low inter-class variability, whereas domain-specific pretraining is essential to achieve robust and clinically reliable representations in fetal brain ultrasound imaging.
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。
