发表机构
Università degli Studi di Milano; Politecnico di Milano; Human Technopole; Otto von Guericke University Magdeburg（米兰大学; 米兰理工大学; 人类技术中心; 奥托·冯·格里克大学马格德堡）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
MIRTO协议通过明确配准、阈值和评估管道选择，在15,552种管道上评估脑MRI无监督异常分割，揭示轴顺序错误大幅降低体素AUROC，且Dice优势在相等假阳性负担下消失。
AI 中文摘要
无监督异常检测（UAD）方法在脑MRI上的排名通常基于单一分数，然而该分数依赖于一些很少被报告的决策：每个异常图如何与参考图像对齐、阈值如何以及在哪些数据上设定，以及使用了何种假阳性预算、指标、聚合方式和病灶定义。我们提出了MIRTO，一种评估协议，使这些选择明确化并衡量其影响。它通过配准检查和已知功效的无标签诊断来门控每次比较的几何形状，仅基于验证数据设定阈值，并报告测试集上实际实现的假阳性体积，在15,552条可辩护的评估管道上重复每次比较，并附上带有多重性控制的配对受试者自助法置信区间。将MIRTO应用于在相同健康数据上训练并在312个BraTS 2020受试者上测试的四种UAD方法，结果显示，存储图与参考之间的轴顺序不匹配使扩散模型的体素AUROC从0.873降至0.583，而其切片级AUROC几乎不变。在每个指标内，该方法解释了体素AUROC和AUPRC中至少0.95的方差，以及Dice中0.77的方差，但在病灶敏感性中仅解释了0.14的方差，其中病灶定义和命中标准占主导地位。在验证阈值下显著的Dice优势在相等的实际假阳性负担下消失，且一个精确的恒等式将其归因于阈值转移。对REFLECT潜在聚合的无训练更改在相等负担下将Dice提高了0.052。九个假设根据明确标准进行了检验；由于同一队列用于开发该协议，所有推断均为探索性的。
英文摘要
Unsupervised anomaly detection (UAD) methods for brain MRI are ranked by a single score, yet that score rests on choices that are rarely reported: how each anomaly map is aligned with the reference, how and on which data the threshold is set, and which false-positive budget, metric, aggregation and lesion definition are used. We present MIRTO, an evaluation protocol that makes these choices explicit and measures their effect. It gates the geometry of every comparison with a registration check and label-free diagnostics of known power, sets thresholds on validation data alone and reports the false-positive volume actually realised on test, repeats each comparison over 15,552 defensible evaluation pipelines, and attaches paired subject-bootstrap intervals with multiplicity control. Applied to four UAD methods trained on the same healthy data and tested on 312 BraTS 2020 subjects, MIRTO showed that an axis-order mismatch between stored maps and the reference lowered a diffusion model's voxel AUROC from 0.873 to 0.583 whilst barely moving its slice-level AUROC. Within each metric, the method explained at least 0.95 of the variance in voxel AUROC and AUPRC and 0.77 in Dice, but only 0.14 in lesion sensitivity, where the lesion definition and hit criterion dominated. A Dice advantage that was significant at validation thresholds vanished at equal realised false-positive burden, and an exact identity attributes it to threshold transfer. A training-free change to REFLECT's latent aggregation raised Dice at equal burden by 0.052. Nine hypotheses were tested against explicit criteria; because the same cohort served to develop the protocol, all inference is exploratory.
