Biomedical Engineering Graduate Program, University of Calgary, Calgary, AB, Canada.
Department of Pediatrics, Department of Community Health Sciences, Alberta Children's Hospital Research Institute, University of Calgary, Calgary, AB, Canada.
Artif Intell Med. 2022 Dec;134:102425. doi: 10.1016/j.artmed.2022.102425. Epub 2022 Oct 20.
Many genetic syndromes are associated with distinctive facial features. Several computer-assisted methods have been proposed that make use of facial features for syndrome diagnosis. Training supervised classifiers, the most common approach for this purpose, requires large, comprehensive, and difficult to collect databases of syndromic facial images. In this work, we use unsupervised, normalizing flow-based manifold and density estimation models trained entirely on unaffected subjects to detect syndromic 3D faces as statistical outliers. Furthermore, we demonstrate a general, user-friendly, gradient-based interpretability mechanism that enables clinicians and patients to understand model inferences. 3D facial surface scans of 2471 unaffected subjects and 1629 syndromic subjects representing 262 different genetic syndromes were used to train and evaluate the models. The flow-based models outperformed unsupervised comparison methods, with the best model achieving an ROC-AUC of 86.3% on a challenging, age and sex diverse data set. In addition to highlighting the viability of outlier-based syndrome screening tools, our methods generalize and extend previously proposed outlier scores for 3D face-based syndrome detection, resulting in improved performance for unsupervised syndrome detection.
许多遗传综合征都与独特的面部特征有关。已经提出了几种计算机辅助方法，这些方法利用面部特征进行综合征诊断。为此目的，训练有监督的分类器是最常用的方法，但这种方法需要大型、全面且难以收集的综合征面部图像数据库。在这项工作中，我们使用无监督的、基于归一化流形和密度估计的模型，这些模型完全是在未受影响的受试者身上训练的，用于检测综合征的 3D 面部作为统计异常值。此外，我们展示了一种通用的、用户友好的基于梯度的可解释性机制，使临床医生和患者能够理解模型的推断。使用 2471 名未受影响的受试者和 1629 名代表 262 种不同遗传综合征的综合征受试者的 3D 面部表面扫描来训练和评估模型。基于流的模型优于无监督的比较方法，最佳模型在具有挑战性的、年龄和性别多样化的数据集上的 ROC-AUC 达到 86.3%。除了突出基于异常值的综合征筛选工具的可行性外，我们的方法还推广和扩展了先前提出的基于 3D 面部的综合征检测的异常值评分，从而提高了无监督的综合征检测性能。
