发表机构
Institute for Genomic Statistics and Bioinformatics; University Hospital Bonn（基因组统计与生物信息学研究所; 波恩大学医院）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
该研究针对罕见遗传病优先级排序的面部表型检索，提出推理时多级证据聚合框架，无需修改GestaltMatcher-Arc编码器，在GMDB数据集上显著提升了各子集的Top-1检索准确率。
AI 中文摘要
AI辅助面部表型分析可通过从GestaltMatcher数据库（GMDB）等面部图像参考数据库中检索视觉相似的确诊病例，支持罕见遗传病的优先级排序。现有基于GestaltMatcher的检索框架在面部表型嵌入空间中将每个测试图像与单个图库图像进行比较，但这种逐点式方案未充分利用可用证据，因为患者可能有多张图像，且疾病可能由多个确诊图库患者代表。我们提出一种推理时多级证据聚合框架，无需修改底层的GestaltMatcher-Arc编码器即可提升面部表型检索性能。该框架整合了同一患者多张图像的嵌入级患者聚合、患者加权疾病质心，以及混合个体-质心评分，以融合测试患者观测、疾病级图库证据和局部近邻证据。我们在GMDB v1.1.4上评估了该方法，涵盖训练时出现的疾病（GMDB-Freq）、未见过的疾病（GMDB-Rare）和多图像患者子集，使用包含GMDB-Freq和GMDB-Rare疾病的统一图库。多级证据聚合在所有评估子集上提升了按疾病均值计算的Top-N检索准确率：GMDB-Freq的Top-1准确率从38.52%提升至48.82%，GMDB-Rare的Top-1准确率从19.38%提升至23.79%；多图像子集方面，GMDB-Multi-Freq的Top-1准确率从46.12%提升至60.94%，GMDB-Multi-Rare的Top-1准确率从18.54%提升至26.71%。这些结果表明，推理时聚合无需重新训练编码器即可改进下一代面部表型检索，推动从孤立的单图像匹配转向用于罕见疾病优先级排序的患者与疾病证据的多级聚合。
英文摘要
AI-assisted facial phenotyping supports rare genetic disorder prioritization by retrieving visually similar diagnosed cases from facial image reference databases such as the GestaltMatcher Database (GMDB). Existing GestaltMatcher-based retrieval frameworks compare each test image with individual gallery images in a facial phenotype embedding space. However, this pointwise formulation does not fully exploit available evidence, because patients may have multiple images and disorders may be represented by multiple diagnosed gallery patients. We propose an inference-time multi-level evidence aggregation framework that improves facial phenotype retrieval without modifying the underlying GestaltMatcher-Arc encoder. The framework combines embedding-level patient aggregation of multiple images from the same individual, patient-weighted disorder centroids, and hybrid individual-centroid scoring to integrate test-patient observations, disorder-level gallery evidence, and local nearest-neighbor evidence. We evaluated the approach on GMDB v1.1.4 across disorders represented during training (GMDB-Freq), unseen disorders (GMDB-Rare), and multi-image patient subsets, using a unified gallery containing both GMDB-Freq and GMDB-Rare disorders. Multi-level evidence aggregation improved mean per-disorder top-$N$ retrieval accuracy across all evaluation subsets. Top-1 accuracy increased from 38.52% to 48.82% on GMDB-Freq and from 19.38% to 23.79% on GMDB-Rare. On multi-image subsets, top-1 accuracy increased from 46.12% to 60.94% on GMDB-Multi-Freq and from 18.54% to 26.71% on GMDB-Multi-Rare. These findings show that inference-time aggregation can improve next-generation facial phenotype retrieval without retraining the encoder, supporting a shift from isolated single-image matching toward multi-level aggregation of patient and disorder evidence for rare-disorder prioritization.
Commentsincluding supplementary notes: 32 pages, 16 figures. Preprint submitted to journal for peer-review
