Learning Prostate Anatomy at Test Time for Cancer Detection in Micro-Ultrasound
Learning Prostate Anatomy at Test Time for Cancer Detection in Micro-Ultrasound
在测试时学习前列腺解剖结构以用于微超声癌症检测
Abstract: Domain shift across clinical centers using different imaging hardware or acquisition protocols remains a fundamental barrier to deploying deep learning models for prostate cancer (PCa) detection. 摘要： 在使用不同成像硬件或采集协议的临床中心之间，领域偏移（Domain shift）仍然是部署深度学习模型进行前列腺癌（PCa）检测的根本障碍。
Existing test-time adaptation (TTA) methods address distribution shift through entropy minimization or augmentation-based self-supervision, correcting for statistical differences in image appearance but ignoring the anatomical structure of the target domain. 现有的测试时适应（TTA）方法通过熵最小化或基于增强的自监督学习来解决分布偏移问题，虽然纠正了图像外观上的统计差异，但忽略了目标领域的解剖结构。
We propose ANT, a segmentation-guided TTA framework that adapts a pretrained cancer detection encoder to the target domain by solving an auxiliary prostate segmentation task at test time, supervised by pseudo-masks from a frozen pretrained segmentation network. 我们提出了 ANT，这是一个分割引导的 TTA 框架。它通过在测试时解决辅助性的前列腺分割任务，将预训练的癌症检测编码器适应到目标领域，并由冻结的预训练分割网络生成的伪掩码（pseudo-masks）进行监督。
By aligning encoder representations to prostate anatomy in the target domain, ANT corrects domain-specific feature drift while preserving cancer-discriminative structure. 通过将编码器表示与目标领域中的前列腺解剖结构对齐，ANT 在纠正特定领域特征漂移的同时，保留了具有癌症判别力的结构。
The model was trained on 693 patients imaged with an earlier-generation micro-ultrasound scanner in a multi-center clinical trial, and evaluated on 118 patients acquired with a newer-generation system across two centers in another clinical trial. 该模型在多中心临床试验中，使用早期微超声扫描仪对 693 名患者进行成像并训练；随后在另一项临床试验中，使用新一代系统对两个中心的 118 名患者进行了评估。
Under a leave-one-center-out protocol with identical evaluation conditions across all methods, ANT improves mean AUC by 2.9% and 3.6% at the biopsy-core and patient levels, respectively, over no adaptation, outperforming TTA baselines. 在所有方法采用相同评估条件的“留一中心法”（leave-one-center-out）协议下，与不进行适应相比，ANT 在活检核心（biopsy-core）和患者层面的平均 AUC 分别提高了 2.9% 和 3.6%，性能优于现有的 TTA 基准方法。
Code is available at: this https URL. 代码地址：点击此处访问。