Dong Xu, Ph.D.

Professor & Director of IT Program，Department of Electrical Engineering and Computer Science, Christopher S. Bond Life Sciences Center, University of Missouri

许东

美国密苏里大学 James C. Dowell 荣誉教授、美国密苏里大学计算机科学系主任、教授，美国密苏里大学哥伦比亚分校电器工程与计算机系教授。获得美国“最杰出研究与开发100人奖励（国际2001R&D100Award）”，联邦实验室联合会最杰出技术转让奖，国际智能生命医学协会杰出成就奖。长期从事蛋白质结构和功能、DNA序列的生物信息学研究。主要包括蛋白质结构的预测和建模，以及高通量数据分析。

Applications of Deep Learning in Biomedical Informatics

深度学习在生物医学信息学中的应用

Deep learning has made remarkable improvements on many computational problems and presents a new opportunity for biomedical informatics. The growing amount of biological data also allows deep learning to generate robust models. In this talk, I will present several of our successful deep-learning applications in biomedical informatics, including protein localization prediction, protein post-translational modification prediction, genotype-phenotype relationship, and image understanding of biomedical figures. These applications utilized a broad spectrum of deep learning methods, including Deep Neural Network (DNN), Convolutional Neural Network (CNN), Recurrent neural network (RNN), Generative Adversarial Network (GAN), Capsule Network and Graph Neural Network (GNN). Various network architectures are also explored, including residual network, inception network, dense network, Mask-RCNN, etc. Some of these applications represent novel formulations of the problems, while others significantly improved the performance over the previous methods. These studies also addressed some important deep-learning issues, such as handling small data, using transfer learning to pretrain models, and making the models transparent and explainable.