报告人：Yu-Ping Wang

报告题目：Integration of brain imaging and genomics with interpretable multimodal collaborative learning

时间：2025年12月22日 10:00-11:30

地点：数学楼2-1室

报告摘要：

Recent years have witnessed the convergence of multiscale and multimodal brain imaging and omics techniques, showing great promise for systematic and precision medicine. In the meantime, they bring significant data analysis challenges when integrating and mining these large volumes of heterogeneous datasets. In this work, we first introduce a linear collaborative learning model to combine both regression and correlation analysis such as canonical correlation analysis (CCA). To further capture complex interactions both within and across modalities, we develop an interpretable multimodal deep learning-based integration model to perform heterogeneous data integration and result interpretation simultaneously. The proposed model can generate interpretable activation maps to quantify the contribution of imaging or omics features. Moreover, the estimated activation maps are class-specific, which can therefore facilitate the identification of biomarkers. Finally, we apply and validate the  model in the study of brain development with integrative analysis of multi-modal brain imaging and genomics data. We demonstrate its successful application to both the classification of cognitive function groups and the discovery of underlying genetic mechanisms.

报告人简介：

Dr. Yu-Ping Wang，美国杜兰大学科学与工程学院及公共卫生与热带医学院的生物医学工程、生物统计学与数据科学终身教授，杜兰大学生物信息学与基因组学中心、杜兰癌症中心以及杜兰神经科学项目的成员。研究方向包括计算机视觉、信号处理与机器学习在生物医学成像及生物信息学中的应用，已发表期刊论文 260余篇。过去十年间作为首席研究员获得美国国立卫生研究院和美国国家科学基金会超过1400万美元的科研资助。他曾任职于多个学术会议程序委员会及 NSF、NIH评审专家组，并担任J. Neuroscience Methods、IEEE/ACM Trans. Computational Biology and Bioinformatics、IEEE Trans. Medical Imaging,等多个期刊的编委。他是美国生物与医学工程研究院会士，荣获 2022 年杜兰大学融合奖（Tulane Convergence Award）及 2025 年国际智能生物学与医学协会（IAIBM）杰出学者奖。

邀请人：乔琛教授