arXiv 2602.062822026-02-06cs.CVq-bio.QM

一种可解释的基于指纹的视觉Transformer辅助诊断Kabuki和Wiedemann-Steiner综合征

An Interpretable Vision Transformer as a Fingerprint-Based Diagnostic Aid for Kabuki and Wiedemann-Steiner Syndromes

Dept. of Computer Science, Vanderbilt University（范德比尔特大学计算机科学系）；Dept. of Genetics and Molecular Medicine, Landspitali University Hospital（陆斯帕蒂医院遗传学与分子医学系）；Louma G. Laboratory of Epigenetics Research, Faculty of Medicine, University of Iceland（爱沙尼亚大学医学系表观遗传学研究实验室）；McKusick-Nathans Dept. of Genetic Medicine, Johns Hopkins University School of Medicine（约翰霍普金斯大学医学院遗传医学部）；Faculty of Electrical and Computer Engineering University of Iceland（爱沙尼亚大学电气与计算机工程系）

Marilyn Lionts, Arnhildur Tomasdottir, Viktor I. Agustsson, Yuankai Huo, Hans T. Bjornsson, Lotta M. Ellingsen

DOI：: 10.1117/12.3085208

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

本研究提出一种基于视觉Transformer的深度学习模型，利用指纹图像区分Kabuki综合征（KS）和Wiedemann-Steiner综合征（WSS）患者与健康对照，并通过注意力可视化增强可解释性，为罕见遗传病的非侵入性诊断提供新工具。

摘要：

Kabuki综合征（KS）和Wiedemann-Steiner综合征（WSS）是罕见但不同的发育障碍，具有重叠的临床特征，包括神经发育迟缓、生长受限和持续性胎儿指尖垫。尽管基因检测仍是诊断的金标准，但由于基因检测和专业知识获取的障碍，许多KS或WSS患者仍未得到诊断。皮纹异常虽然是几种遗传综合征的既定标志，但在分子检测时代仍是一种未被充分利用的诊断信号。本研究提出一种基于视觉Transformer的深度学习模型，利用指纹图像区分KS和WSS患者与未受影响的对照组以及彼此。我们在三个二分类任务中评估模型性能。在三个分类任务中，模型在对照组vs. KS、对照组vs. WSS和KS vs. WSS上分别达到了0.80、0.73和0.85的AUC分数，相应的F1分数分别为0.71、0.72和0.83。除了分类，我们应用基于注意力的可视化来识别对模型预测最显著的指纹区域，增强了可解释性。总之，这些发现表明存在综合征特异性的指纹特征，证明了基于指纹的人工智能（AI）工具作为一种非侵入性、可解释且可获取的未来诊断辅助手段，用于早期诊断未充分诊断的遗传综合征的可行性。

英文摘要：

Kabuki syndrome (KS) and Wiedemann-Steiner syndrome (WSS) are rare but distinct developmental disorders that share overlapping clinical features, including neurodevelopmental delay, growth restriction, and persistent fetal fingertip pads. While genetic testing remains the diagnostic gold standard, many individuals with KS or WSS remain undiagnosed due to barriers in access to both genetic testing and expertise. Dermatoglyphic anomalies, despite being established hallmarks of several genetic syndromes, remain an underutilized diagnostic signal in the era of molecular testing. This study presents a vision transformer-based deep learning model that leverages fingerprint images to distinguish individuals with KS and WSS from unaffected controls and from one another. We evaluate model performance across three binary classification tasks. Across the three classification tasks, the model achieved AUC scores of 0.80 (control vs. KS), 0.73 (control vs. WSS), and 0.85 (KS vs. WSS), with corresponding F1 scores of 0.71, 0.72, and 0.83, respectively. Beyond classification, we apply attention-based visualizations to identify fingerprint regions most salient to model predictions, enhancing interpretability. Together, these findings suggest the presence of syndrome-specific fingerprint features, demonstrating the feasibility of a fingerprint-based artificial intelligence (AI) tool as a noninvasive, interpretable, and accessible future diagnostic aid for the early diagnosis of underdiagnosed genetic syndromes.

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。